From: Sasha Levin Date: Mon, 7 Jul 2025 04:22:40 +0000 (-0400) Subject: Fixes for 6.15 X-Git-Tag: v5.15.187~56 X-Git-Url: http://git.ipfire.org/cgi-bin/gitweb.cgi?a=commitdiff_plain;h=eb321994dba1733f98ff5ea2a4e0a900ae8cddf0;p=thirdparty%2Fkernel%2Fstable-queue.git Fixes for 6.15 Signed-off-by: Sasha Levin --- diff --git a/queue-6.15/acpica-refuse-to-evaluate-a-method-if-arguments-are-.patch b/queue-6.15/acpica-refuse-to-evaluate-a-method-if-arguments-are-.patch new file mode 100644 index 0000000000..894e708cb1 --- /dev/null +++ b/queue-6.15/acpica-refuse-to-evaluate-a-method-if-arguments-are-.patch @@ -0,0 +1,51 @@ +From fed7a33873c47f72bc10b46b011d0e7bd0d1f74d Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Wed, 18 Jun 2025 14:17:45 +0200 +Subject: ACPICA: Refuse to evaluate a method if arguments are missing + +From: Rafael J. Wysocki + +[ Upstream commit 6fcab2791543924d438e7fa49276d0998b0a069f ] + +As reported in [1], a platform firmware update that increased the number +of method parameters and forgot to update a least one of its callers, +caused ACPICA to crash due to use-after-free. + +Since this a result of a clear AML issue that arguably cannot be fixed +up by the interpreter (it cannot produce missing data out of thin air), +address it by making ACPICA refuse to evaluate a method if the caller +attempts to pass fewer arguments than expected to it. + +Closes: https://github.com/acpica/acpica/issues/1027 [1] +Reported-by: Peter Williams +Signed-off-by: Rafael J. Wysocki +Reviewed-by: Hans de Goede +Tested-by: Hans de Goede # Dell XPS 9640 with BIOS 1.12.0 +Link: https://patch.msgid.link/5909446.DvuYhMxLoT@rjwysocki.net +Signed-off-by: Rafael J. Wysocki +Signed-off-by: Sasha Levin +--- + drivers/acpi/acpica/dsmethod.c | 7 +++++++ + 1 file changed, 7 insertions(+) + +diff --git a/drivers/acpi/acpica/dsmethod.c b/drivers/acpi/acpica/dsmethod.c +index e809c2aed78ae..a232746d150a7 100644 +--- a/drivers/acpi/acpica/dsmethod.c ++++ b/drivers/acpi/acpica/dsmethod.c +@@ -483,6 +483,13 @@ acpi_ds_call_control_method(struct acpi_thread_state *thread, + return_ACPI_STATUS(AE_NULL_OBJECT); + } + ++ if (this_walk_state->num_operands < obj_desc->method.param_count) { ++ ACPI_ERROR((AE_INFO, "Missing argument for method [%4.4s]", ++ acpi_ut_get_node_name(method_node))); ++ ++ return_ACPI_STATUS(AE_AML_UNINITIALIZED_ARG); ++ } ++ + /* Init for new method, possibly wait on method mutex */ + + status = +-- +2.39.5 + diff --git a/queue-6.15/alsa-sb-don-t-allow-changing-the-dma-mode-during-ope.patch b/queue-6.15/alsa-sb-don-t-allow-changing-the-dma-mode-during-ope.patch new file mode 100644 index 0000000000..e337e022dd --- /dev/null +++ b/queue-6.15/alsa-sb-don-t-allow-changing-the-dma-mode-during-ope.patch @@ -0,0 +1,38 @@ +From bd77cc927a9dd1b99f5c0745a43a38ebc86206ed Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Tue, 10 Jun 2025 08:43:19 +0200 +Subject: ALSA: sb: Don't allow changing the DMA mode during operations + +From: Takashi Iwai + +[ Upstream commit ed29e073ba93f2d52832804cabdd831d5d357d33 ] + +When a PCM stream is already running, one shouldn't change the DMA +mode via kcontrol, which may screw up the hardware. Return -EBUSY +instead. + +Link: https://bugzilla.kernel.org/show_bug.cgi?id=218185 +Link: https://patch.msgid.link/20250610064322.26787-1-tiwai@suse.de +Signed-off-by: Takashi Iwai +Signed-off-by: Sasha Levin +--- + sound/isa/sb/sb16_main.c | 3 +++ + 1 file changed, 3 insertions(+) + +diff --git a/sound/isa/sb/sb16_main.c b/sound/isa/sb/sb16_main.c +index 74db115250030..c4930efd44e3a 100644 +--- a/sound/isa/sb/sb16_main.c ++++ b/sound/isa/sb/sb16_main.c +@@ -703,6 +703,9 @@ static int snd_sb16_dma_control_put(struct snd_kcontrol *kcontrol, struct snd_ct + unsigned char nval, oval; + int change; + ++ if (chip->mode & (SB_MODE_PLAYBACK | SB_MODE_CAPTURE)) ++ return -EBUSY; ++ + nval = ucontrol->value.enumerated.item[0]; + if (nval > 2) + return -EINVAL; +-- +2.39.5 + diff --git a/queue-6.15/alsa-sb-force-to-disable-dmas-once-when-dma-mode-is-.patch b/queue-6.15/alsa-sb-force-to-disable-dmas-once-when-dma-mode-is-.patch new file mode 100644 index 0000000000..5c4746b63b --- /dev/null +++ b/queue-6.15/alsa-sb-force-to-disable-dmas-once-when-dma-mode-is-.patch @@ -0,0 +1,41 @@ +From 1b750ced45034a39862ae0a44610a6ef837e0433 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Tue, 10 Jun 2025 08:43:20 +0200 +Subject: ALSA: sb: Force to disable DMAs once when DMA mode is changed + +From: Takashi Iwai + +[ Upstream commit 4c267ae2ef349639b4d9ebf00dd28586a82fdbe6 ] + +When the DMA mode is changed on the (still real!) SB AWE32 after +playing a stream and closing, the previous DMA setup was still +silently kept, and it can confuse the hardware, resulting in the +unexpected noises. As a workaround, enforce the disablement of DMA +setups when the DMA setup is changed by the kcontrol. + +https://bugzilla.kernel.org/show_bug.cgi?id=218185 +Link: https://patch.msgid.link/20250610064322.26787-2-tiwai@suse.de +Signed-off-by: Takashi Iwai +Signed-off-by: Sasha Levin +--- + sound/isa/sb/sb16_main.c | 4 ++++ + 1 file changed, 4 insertions(+) + +diff --git a/sound/isa/sb/sb16_main.c b/sound/isa/sb/sb16_main.c +index c4930efd44e3a..5a083eecaa6b9 100644 +--- a/sound/isa/sb/sb16_main.c ++++ b/sound/isa/sb/sb16_main.c +@@ -714,6 +714,10 @@ static int snd_sb16_dma_control_put(struct snd_kcontrol *kcontrol, struct snd_ct + change = nval != oval; + snd_sb16_set_dma_mode(chip, nval); + spin_unlock_irqrestore(&chip->reg_lock, flags); ++ if (change) { ++ snd_dma_disable(chip->dma8); ++ snd_dma_disable(chip->dma16); ++ } + return change; + } + +-- +2.39.5 + diff --git a/queue-6.15/amd-xgbe-align-cl37-an-sequence-as-per-databook.patch b/queue-6.15/amd-xgbe-align-cl37-an-sequence-as-per-databook.patch new file mode 100644 index 0000000000..dd668c9f80 --- /dev/null +++ b/queue-6.15/amd-xgbe-align-cl37-an-sequence-as-per-databook.patch @@ -0,0 +1,90 @@ +From 555f25ccd495095c30ac8091a5f7886df8855d98 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Tue, 1 Jul 2025 00:56:36 +0530 +Subject: amd-xgbe: align CL37 AN sequence as per databook + +From: Raju Rangoju + +[ Upstream commit 42fd432fe6d320323215ebdf4de4d0d7e56e6792 ] + +Update the Clause 37 Auto-Negotiation implementation to properly align +with the PCS hardware specifications: +- Fix incorrect bit settings in Link Status and Link Duplex fields +- Implement missing sequence steps 2 and 7 + +These changes ensure CL37 auto-negotiation protocol follows the exact +sequence patterns as specified in the hardware databook. + +Fixes: 1bf40ada6290 ("amd-xgbe: Add support for clause 37 auto-negotiation") +Signed-off-by: Raju Rangoju +Link: https://patch.msgid.link/20250630192636.3838291-1-Raju.Rangoju@amd.com +Signed-off-by: Jakub Kicinski +Signed-off-by: Sasha Levin +--- + drivers/net/ethernet/amd/xgbe/xgbe-common.h | 2 ++ + drivers/net/ethernet/amd/xgbe/xgbe-mdio.c | 9 +++++++++ + drivers/net/ethernet/amd/xgbe/xgbe.h | 4 ++-- + 3 files changed, 13 insertions(+), 2 deletions(-) + +diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-common.h b/drivers/net/ethernet/amd/xgbe/xgbe-common.h +index 3b70f67376331..aa25a8a0a106f 100644 +--- a/drivers/net/ethernet/amd/xgbe/xgbe-common.h ++++ b/drivers/net/ethernet/amd/xgbe/xgbe-common.h +@@ -1373,6 +1373,8 @@ + #define MDIO_VEND2_CTRL1_SS13 BIT(13) + #endif + ++#define XGBE_VEND2_MAC_AUTO_SW BIT(9) ++ + /* MDIO mask values */ + #define XGBE_AN_CL73_INT_CMPLT BIT(0) + #define XGBE_AN_CL73_INC_LINK BIT(1) +diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-mdio.c b/drivers/net/ethernet/amd/xgbe/xgbe-mdio.c +index 07f4f3418d018..3316c719f9f8c 100644 +--- a/drivers/net/ethernet/amd/xgbe/xgbe-mdio.c ++++ b/drivers/net/ethernet/amd/xgbe/xgbe-mdio.c +@@ -375,6 +375,10 @@ static void xgbe_an37_set(struct xgbe_prv_data *pdata, bool enable, + reg |= MDIO_VEND2_CTRL1_AN_RESTART; + + XMDIO_WRITE(pdata, MDIO_MMD_VEND2, MDIO_CTRL1, reg); ++ ++ reg = XMDIO_READ(pdata, MDIO_MMD_VEND2, MDIO_PCS_DIG_CTRL); ++ reg |= XGBE_VEND2_MAC_AUTO_SW; ++ XMDIO_WRITE(pdata, MDIO_MMD_VEND2, MDIO_PCS_DIG_CTRL, reg); + } + + static void xgbe_an37_restart(struct xgbe_prv_data *pdata) +@@ -1003,6 +1007,11 @@ static void xgbe_an37_init(struct xgbe_prv_data *pdata) + + netif_dbg(pdata, link, pdata->netdev, "CL37 AN (%s) initialized\n", + (pdata->an_mode == XGBE_AN_MODE_CL37) ? "BaseX" : "SGMII"); ++ ++ reg = XMDIO_READ(pdata, MDIO_MMD_AN, MDIO_CTRL1); ++ reg &= ~MDIO_AN_CTRL1_ENABLE; ++ XMDIO_WRITE(pdata, MDIO_MMD_AN, MDIO_CTRL1, reg); ++ + } + + static void xgbe_an73_init(struct xgbe_prv_data *pdata) +diff --git a/drivers/net/ethernet/amd/xgbe/xgbe.h b/drivers/net/ethernet/amd/xgbe/xgbe.h +index ed5d43c16d0e2..7526a0906b391 100644 +--- a/drivers/net/ethernet/amd/xgbe/xgbe.h ++++ b/drivers/net/ethernet/amd/xgbe/xgbe.h +@@ -292,12 +292,12 @@ + #define XGBE_LINK_TIMEOUT 5 + #define XGBE_KR_TRAINING_WAIT_ITER 50 + +-#define XGBE_SGMII_AN_LINK_STATUS BIT(1) ++#define XGBE_SGMII_AN_LINK_DUPLEX BIT(1) + #define XGBE_SGMII_AN_LINK_SPEED (BIT(2) | BIT(3)) + #define XGBE_SGMII_AN_LINK_SPEED_10 0x00 + #define XGBE_SGMII_AN_LINK_SPEED_100 0x04 + #define XGBE_SGMII_AN_LINK_SPEED_1000 0x08 +-#define XGBE_SGMII_AN_LINK_DUPLEX BIT(4) ++#define XGBE_SGMII_AN_LINK_STATUS BIT(4) + + /* ECC correctable error notification window (seconds) */ + #define XGBE_ECC_LIMIT 60 +-- +2.39.5 + diff --git a/queue-6.15/amd-xgbe-do-not-double-read-link-status.patch b/queue-6.15/amd-xgbe-do-not-double-read-link-status.patch new file mode 100644 index 0000000000..1339179170 --- /dev/null +++ b/queue-6.15/amd-xgbe-do-not-double-read-link-status.patch @@ -0,0 +1,94 @@ +From 15584bccf859c56fb489d880a2f6c064cf0e55bf Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Tue, 1 Jul 2025 12:20:16 +0530 +Subject: amd-xgbe: do not double read link status + +From: Raju Rangoju + +[ Upstream commit 16ceda2ef683a50cd0783006c0504e1931cd8879 ] + +The link status is latched low so that momentary link drops +can be detected. Always double-reading the status defeats this +design feature. Only double read if link was already down + +This prevents unnecessary duplicate readings of the link status. + +Fixes: 4f3b20bfbb75 ("amd-xgbe: add support for rx-adaptation") +Signed-off-by: Raju Rangoju +Reviewed-by: Simon Horman +Link: https://patch.msgid.link/20250701065016.4140707-1-Raju.Rangoju@amd.com +Signed-off-by: Jakub Kicinski +Signed-off-by: Sasha Levin +--- + drivers/net/ethernet/amd/xgbe/xgbe-mdio.c | 4 ++++ + drivers/net/ethernet/amd/xgbe/xgbe-phy-v2.c | 24 +++++++++++++-------- + 2 files changed, 19 insertions(+), 9 deletions(-) + +diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-mdio.c b/drivers/net/ethernet/amd/xgbe/xgbe-mdio.c +index 3316c719f9f8c..ed76a8df6ec6e 100644 +--- a/drivers/net/ethernet/amd/xgbe/xgbe-mdio.c ++++ b/drivers/net/ethernet/amd/xgbe/xgbe-mdio.c +@@ -1413,6 +1413,10 @@ static void xgbe_phy_status(struct xgbe_prv_data *pdata) + + pdata->phy.link = pdata->phy_if.phy_impl.link_status(pdata, + &an_restart); ++ /* bail out if the link status register read fails */ ++ if (pdata->phy.link < 0) ++ return; ++ + if (an_restart) { + xgbe_phy_config_aneg(pdata); + goto adjust_link; +diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-phy-v2.c b/drivers/net/ethernet/amd/xgbe/xgbe-phy-v2.c +index 268399dfcf22f..32e633d113484 100644 +--- a/drivers/net/ethernet/amd/xgbe/xgbe-phy-v2.c ++++ b/drivers/net/ethernet/amd/xgbe/xgbe-phy-v2.c +@@ -2855,8 +2855,7 @@ static bool xgbe_phy_valid_speed(struct xgbe_prv_data *pdata, int speed) + static int xgbe_phy_link_status(struct xgbe_prv_data *pdata, int *an_restart) + { + struct xgbe_phy_data *phy_data = pdata->phy_data; +- unsigned int reg; +- int ret; ++ int reg, ret; + + *an_restart = 0; + +@@ -2890,11 +2889,20 @@ static int xgbe_phy_link_status(struct xgbe_prv_data *pdata, int *an_restart) + return 0; + } + +- /* Link status is latched low, so read once to clear +- * and then read again to get current state +- */ +- reg = XMDIO_READ(pdata, MDIO_MMD_PCS, MDIO_STAT1); + reg = XMDIO_READ(pdata, MDIO_MMD_PCS, MDIO_STAT1); ++ if (reg < 0) ++ return reg; ++ ++ /* Link status is latched low so that momentary link drops ++ * can be detected. If link was already down read again ++ * to get the latest state. ++ */ ++ ++ if (!pdata->phy.link && !(reg & MDIO_STAT1_LSTATUS)) { ++ reg = XMDIO_READ(pdata, MDIO_MMD_PCS, MDIO_STAT1); ++ if (reg < 0) ++ return reg; ++ } + + if (pdata->en_rx_adap) { + /* if the link is available and adaptation is done, +@@ -2913,9 +2921,7 @@ static int xgbe_phy_link_status(struct xgbe_prv_data *pdata, int *an_restart) + xgbe_phy_set_mode(pdata, phy_data->cur_mode); + } + +- /* check again for the link and adaptation status */ +- reg = XMDIO_READ(pdata, MDIO_MMD_PCS, MDIO_STAT1); +- if ((reg & MDIO_STAT1_LSTATUS) && pdata->rx_adapt_done) ++ if (pdata->rx_adapt_done) + return 1; + } else if (reg & MDIO_STAT1_LSTATUS) + return 1; +-- +2.39.5 + diff --git a/queue-6.15/aoe-defer-rexmit-timer-downdev-work-to-workqueue.patch b/queue-6.15/aoe-defer-rexmit-timer-downdev-work-to-workqueue.patch new file mode 100644 index 0000000000..720adbc267 --- /dev/null +++ b/queue-6.15/aoe-defer-rexmit-timer-downdev-work-to-workqueue.patch @@ -0,0 +1,94 @@ +From 23ba2e2a54068d81f0d09fca41bad3cb113a2ba9 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Tue, 10 Jun 2025 17:06:00 +0000 +Subject: aoe: defer rexmit timer downdev work to workqueue + +From: Justin Sanders + +[ Upstream commit cffc873d68ab09a0432b8212008c5613f8a70a2c ] + +When aoe's rexmit_timer() notices that an aoe target fails to respond to +commands for more than aoe_deadsecs, it calls aoedev_downdev() which +cleans the outstanding aoe and block queues. This can involve sleeping, +such as in blk_mq_freeze_queue(), which should not occur in irq context. + +This patch defers that aoedev_downdev() call to the aoe device's +workqueue. + +Link: https://bugzilla.kernel.org/show_bug.cgi?id=212665 +Signed-off-by: Justin Sanders +Link: https://lore.kernel.org/r/20250610170600.869-2-jsanders.devel@gmail.com +Tested-By: Valentin Kleibel +Signed-off-by: Jens Axboe +Signed-off-by: Sasha Levin +--- + drivers/block/aoe/aoe.h | 1 + + drivers/block/aoe/aoecmd.c | 8 ++++++-- + drivers/block/aoe/aoedev.c | 5 ++++- + 3 files changed, 11 insertions(+), 3 deletions(-) + +diff --git a/drivers/block/aoe/aoe.h b/drivers/block/aoe/aoe.h +index 749ae1246f4cf..d35caa3c69e15 100644 +--- a/drivers/block/aoe/aoe.h ++++ b/drivers/block/aoe/aoe.h +@@ -80,6 +80,7 @@ enum { + DEVFL_NEWSIZE = (1<<6), /* need to update dev size in block layer */ + DEVFL_FREEING = (1<<7), /* set when device is being cleaned up */ + DEVFL_FREED = (1<<8), /* device has been cleaned up */ ++ DEVFL_DEAD = (1<<9), /* device has timed out of aoe_deadsecs */ + }; + + enum { +diff --git a/drivers/block/aoe/aoecmd.c b/drivers/block/aoe/aoecmd.c +index 92b06d1de4cc7..6c94cfd1c480e 100644 +--- a/drivers/block/aoe/aoecmd.c ++++ b/drivers/block/aoe/aoecmd.c +@@ -754,7 +754,7 @@ rexmit_timer(struct timer_list *timer) + + utgts = count_targets(d, NULL); + +- if (d->flags & DEVFL_TKILL) { ++ if (d->flags & (DEVFL_TKILL | DEVFL_DEAD)) { + spin_unlock_irqrestore(&d->lock, flags); + return; + } +@@ -786,7 +786,8 @@ rexmit_timer(struct timer_list *timer) + * to clean up. + */ + list_splice(&flist, &d->factive[0]); +- aoedev_downdev(d); ++ d->flags |= DEVFL_DEAD; ++ queue_work(aoe_wq, &d->work); + goto out; + } + +@@ -898,6 +899,9 @@ aoecmd_sleepwork(struct work_struct *work) + { + struct aoedev *d = container_of(work, struct aoedev, work); + ++ if (d->flags & DEVFL_DEAD) ++ aoedev_downdev(d); ++ + if (d->flags & DEVFL_GDALLOC) + aoeblk_gdalloc(d); + +diff --git a/drivers/block/aoe/aoedev.c b/drivers/block/aoe/aoedev.c +index 8c18034cb3d69..9b66f8ebce50d 100644 +--- a/drivers/block/aoe/aoedev.c ++++ b/drivers/block/aoe/aoedev.c +@@ -200,8 +200,11 @@ aoedev_downdev(struct aoedev *d) + struct list_head *head, *pos, *nx; + struct request *rq, *rqnext; + int i; ++ unsigned long flags; + +- d->flags &= ~DEVFL_UP; ++ spin_lock_irqsave(&d->lock, flags); ++ d->flags &= ~(DEVFL_UP | DEVFL_DEAD); ++ spin_unlock_irqrestore(&d->lock, flags); + + /* clean out active and to-be-retransmitted buffers */ + for (i = 0; i < NFACTIVE; i++) { +-- +2.39.5 + diff --git a/queue-6.15/arm64-dts-apple-drop-address-size-cells-from-spi-nor.patch b/queue-6.15/arm64-dts-apple-drop-address-size-cells-from-spi-nor.patch new file mode 100644 index 0000000000..d51d4f5a22 --- /dev/null +++ b/queue-6.15/arm64-dts-apple-drop-address-size-cells-from-spi-nor.patch @@ -0,0 +1,39 @@ +From 6d594011a3dd1e2094d96f66feff730dc887b6d2 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Tue, 10 Jun 2025 19:19:24 +0000 +Subject: arm64: dts: apple: Drop {address,size}-cells from SPI NOR + +From: Sven Peter + +[ Upstream commit 811a909978bf59caa25359e0aca4e30500dcff26 ] + +Fix the following warning by dropping #{address,size}-cells from the SPI +NOR node which only has a single child node without reg property: + +spi1-nvram.dtsi:19.10-38.4: Warning (avoid_unnecessary_addr_size): /soc/spi@235104000/flash@0: unnecessary #address-cells/#size-cells without "ranges", "dma-ranges" or child "reg" property + +Fixes: 3febe9de5ca5 ("arm64: dts: apple: Add SPI NOR nvram partition to all devices") +Reviewed-by: Janne Grunau +Link: https://lore.kernel.org/r/20250610-apple-dts-warnings-v1-1-70b53e8108a0@kernel.org +Signed-off-by: Sven Peter +Signed-off-by: Sasha Levin +--- + arch/arm64/boot/dts/apple/spi1-nvram.dtsi | 2 -- + 1 file changed, 2 deletions(-) + +diff --git a/arch/arm64/boot/dts/apple/spi1-nvram.dtsi b/arch/arm64/boot/dts/apple/spi1-nvram.dtsi +index 3df2fd3993b52..9740fbf200f0b 100644 +--- a/arch/arm64/boot/dts/apple/spi1-nvram.dtsi ++++ b/arch/arm64/boot/dts/apple/spi1-nvram.dtsi +@@ -20,8 +20,6 @@ flash@0 { + compatible = "jedec,spi-nor"; + reg = <0x0>; + spi-max-frequency = <25000000>; +- #address-cells = <1>; +- #size-cells = <1>; + + partitions { + compatible = "fixed-partitions"; +-- +2.39.5 + diff --git a/queue-6.15/arm64-dts-apple-move-touchbar-mipi-address-size-cell.patch b/queue-6.15/arm64-dts-apple-move-touchbar-mipi-address-size-cell.patch new file mode 100644 index 0000000000..63d4b4475a --- /dev/null +++ b/queue-6.15/arm64-dts-apple-move-touchbar-mipi-address-size-cell.patch @@ -0,0 +1,84 @@ +From 9c45f8f78d4568947e256022533520f14b1e9bf9 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Wed, 11 Jun 2025 15:18:53 +0000 +Subject: arm64: dts: apple: Move touchbar mipi {address,size}-cells from dtsi + to dts + +From: Sven Peter + +[ Upstream commit 08a0d93c353bd55de8b5fb77b464d89425be0215 ] + +Move the {address,size}-cells property from the (disabled) touchbar screen +mipi node inside the dtsi file to the model-specific dts file where it's +enabled to fix the following W=1 warnings: + +t8103.dtsi:404.34-433.5: Warning (avoid_unnecessary_addr_size): /soc/dsi@228600000: unnecessary #address-cells/#size-cells without "ranges", "dma-ranges" or child "reg" property +t8112.dtsi:419.34-448.5: Warning (avoid_unnecessary_addr_size): /soc/dsi@228600000: unnecessary #address-cells/#size-cells without "ranges", "dma-ranges" or child "reg" property + +Fixes: 7275e795e520 ("arm64: dts: apple: Add touchbar screen nodes") +Reviewed-by: Janne Grunau +Link: https://lore.kernel.org/r/20250611-display-pipe-mipi-warning-v1-1-bd80ba2c0eea@kernel.org +Signed-off-by: Sven Peter +Signed-off-by: Sasha Levin +--- + arch/arm64/boot/dts/apple/t8103-j293.dts | 2 ++ + arch/arm64/boot/dts/apple/t8103.dtsi | 2 -- + arch/arm64/boot/dts/apple/t8112-j493.dts | 2 ++ + arch/arm64/boot/dts/apple/t8112.dtsi | 2 -- + 4 files changed, 4 insertions(+), 4 deletions(-) + +diff --git a/arch/arm64/boot/dts/apple/t8103-j293.dts b/arch/arm64/boot/dts/apple/t8103-j293.dts +index e2d9439397f71..5b3c42e9f0e67 100644 +--- a/arch/arm64/boot/dts/apple/t8103-j293.dts ++++ b/arch/arm64/boot/dts/apple/t8103-j293.dts +@@ -100,6 +100,8 @@ dfr_mipi_out_panel: endpoint@0 { + + &displaydfr_mipi { + status = "okay"; ++ #address-cells = <1>; ++ #size-cells = <0>; + + dfr_panel: panel@0 { + compatible = "apple,j293-summit", "apple,summit"; +diff --git a/arch/arm64/boot/dts/apple/t8103.dtsi b/arch/arm64/boot/dts/apple/t8103.dtsi +index 97b6a067394e3..229b10efaab9e 100644 +--- a/arch/arm64/boot/dts/apple/t8103.dtsi ++++ b/arch/arm64/boot/dts/apple/t8103.dtsi +@@ -404,8 +404,6 @@ displaydfr_mipi: dsi@228600000 { + compatible = "apple,t8103-display-pipe-mipi", "apple,h7-display-pipe-mipi"; + reg = <0x2 0x28600000 0x0 0x100000>; + power-domains = <&ps_mipi_dsi>; +- #address-cells = <1>; +- #size-cells = <0>; + status = "disabled"; + + ports { +diff --git a/arch/arm64/boot/dts/apple/t8112-j493.dts b/arch/arm64/boot/dts/apple/t8112-j493.dts +index be86d34c6696c..fb8ad7d4c65a8 100644 +--- a/arch/arm64/boot/dts/apple/t8112-j493.dts ++++ b/arch/arm64/boot/dts/apple/t8112-j493.dts +@@ -63,6 +63,8 @@ dfr_mipi_out_panel: endpoint@0 { + + &displaydfr_mipi { + status = "okay"; ++ #address-cells = <1>; ++ #size-cells = <0>; + + dfr_panel: panel@0 { + compatible = "apple,j493-summit", "apple,summit"; +diff --git a/arch/arm64/boot/dts/apple/t8112.dtsi b/arch/arm64/boot/dts/apple/t8112.dtsi +index d9b966d68e4fa..7488e3850493b 100644 +--- a/arch/arm64/boot/dts/apple/t8112.dtsi ++++ b/arch/arm64/boot/dts/apple/t8112.dtsi +@@ -420,8 +420,6 @@ displaydfr_mipi: dsi@228600000 { + compatible = "apple,t8112-display-pipe-mipi", "apple,h7-display-pipe-mipi"; + reg = <0x2 0x28600000 0x0 0x100000>; + power-domains = <&ps_mipi_dsi>; +- #address-cells = <1>; +- #size-cells = <0>; + status = "disabled"; + + ports { +-- +2.39.5 + diff --git a/queue-6.15/arm64-dts-apple-t8103-fix-pcie-bcm4377-nodename.patch b/queue-6.15/arm64-dts-apple-t8103-fix-pcie-bcm4377-nodename.patch new file mode 100644 index 0000000000..d801e64862 --- /dev/null +++ b/queue-6.15/arm64-dts-apple-t8103-fix-pcie-bcm4377-nodename.patch @@ -0,0 +1,42 @@ +From 0e83346a13b2235ddb2eb71c2b38e4249f7c5279 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Wed, 11 Jun 2025 22:30:31 +0200 +Subject: arm64: dts: apple: t8103: Fix PCIe BCM4377 nodename + +From: Janne Grunau + +[ Upstream commit ac1daa91e9370e3b88ef7826a73d62a4d09e2717 ] + +Fix the following `make dtbs_check` warnings for all t8103 based devices: + +arch/arm64/boot/dts/apple/t8103-j274.dtb: network@0,0: $nodename:0: 'network@0,0' does not match '^wifi(@.*)?$' + from schema $id: http://devicetree.org/schemas/net/wireless/brcm,bcm4329-fmac.yaml# +arch/arm64/boot/dts/apple/t8103-j274.dtb: network@0,0: Unevaluated properties are not allowed ('local-mac-address' was unexpected) + from schema $id: http://devicetree.org/schemas/net/wireless/brcm,bcm4329-fmac.yaml# + +Fixes: bf2c05b619ff ("arm64: dts: apple: t8103: Expose PCI node for the WiFi MAC address") +Signed-off-by: Janne Grunau +Reviewed-by: Sven Peter +Link: https://lore.kernel.org/r/20250611-arm64_dts_apple_wifi-v1-1-fb959d8e1eb4@jannau.net +Signed-off-by: Sven Peter +Signed-off-by: Sasha Levin +--- + arch/arm64/boot/dts/apple/t8103-jxxx.dtsi | 2 +- + 1 file changed, 1 insertion(+), 1 deletion(-) + +diff --git a/arch/arm64/boot/dts/apple/t8103-jxxx.dtsi b/arch/arm64/boot/dts/apple/t8103-jxxx.dtsi +index 8e82231acab59..0c8206156bfef 100644 +--- a/arch/arm64/boot/dts/apple/t8103-jxxx.dtsi ++++ b/arch/arm64/boot/dts/apple/t8103-jxxx.dtsi +@@ -71,7 +71,7 @@ hpm1: usb-pd@3f { + */ + &port00 { + bus-range = <1 1>; +- wifi0: network@0,0 { ++ wifi0: wifi@0,0 { + compatible = "pci14e4,4425"; + reg = <0x10000 0x0 0x0 0x0 0x0>; + /* To be filled by the loader */ +-- +2.39.5 + diff --git a/queue-6.15/asoc-amd-yc-add-quirk-for-msi-bravo-17-d7vf-internal.patch b/queue-6.15/asoc-amd-yc-add-quirk-for-msi-bravo-17-d7vf-internal.patch new file mode 100644 index 0000000000..eedc0e66e7 --- /dev/null +++ b/queue-6.15/asoc-amd-yc-add-quirk-for-msi-bravo-17-d7vf-internal.patch @@ -0,0 +1,42 @@ +From 3a41b4685051965394300c5947bea6da1b338ee1 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Fri, 30 May 2025 02:52:32 +0200 +Subject: ASoC: amd: yc: Add quirk for MSI Bravo 17 D7VF internal mic + +From: Gabriel Santese + +[ Upstream commit ba06528ad5a31923efc24324706116ccd17e12d8 ] + +MSI Bravo 17 (D7VF), like other laptops from the family, +has broken ACPI tables and needs a quirk for internal mic +to work properly. + +Signed-off-by: Gabriel Santese +Link: https://patch.msgid.link/20250530005444.23398-1-santesegabriel@gmail.com +Signed-off-by: Mark Brown +Signed-off-by: Sasha Levin +--- + sound/soc/amd/yc/acp6x-mach.c | 7 +++++++ + 1 file changed, 7 insertions(+) + +diff --git a/sound/soc/amd/yc/acp6x-mach.c b/sound/soc/amd/yc/acp6x-mach.c +index b27966f82c8b6..4f8481c6802b1 100644 +--- a/sound/soc/amd/yc/acp6x-mach.c ++++ b/sound/soc/amd/yc/acp6x-mach.c +@@ -451,6 +451,13 @@ static const struct dmi_system_id yc_acp_quirk_table[] = { + DMI_MATCH(DMI_PRODUCT_NAME, "Bravo 17 D7VEK"), + } + }, ++ { ++ .driver_data = &acp6x_card, ++ .matches = { ++ DMI_MATCH(DMI_BOARD_VENDOR, "Micro-Star International Co., Ltd."), ++ DMI_MATCH(DMI_PRODUCT_NAME, "Bravo 17 D7VF"), ++ } ++ }, + { + .driver_data = &acp6x_card, + .matches = { +-- +2.39.5 + diff --git a/queue-6.15/asoc-amd-yc-update-quirk-data-for-hp-victus.patch b/queue-6.15/asoc-amd-yc-update-quirk-data-for-hp-victus.patch new file mode 100644 index 0000000000..9a837f63e5 --- /dev/null +++ b/queue-6.15/asoc-amd-yc-update-quirk-data-for-hp-victus.patch @@ -0,0 +1,40 @@ +From 3059574182dc623d5892dac9564974c0153fcb6c Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Fri, 13 Jun 2025 07:51:25 -0400 +Subject: ASoC: amd: yc: update quirk data for HP Victus + +From: Raven Black + +[ Upstream commit 13b86ea92ebf0fa587fbadfb8a60ca2e9993203f ] + +Make the internal microphone work on HP Victus laptops. + +Signed-off-by: Raven Black +Link: https://patch.msgid.link/20250613-support-hp-victus-microphone-v1-1-bebc4c3a2041@gmail.com +Signed-off-by: Mark Brown +Signed-off-by: Sasha Levin +--- + sound/soc/amd/yc/acp6x-mach.c | 7 +++++++ + 1 file changed, 7 insertions(+) + +diff --git a/sound/soc/amd/yc/acp6x-mach.c b/sound/soc/amd/yc/acp6x-mach.c +index 4f8481c6802b1..723cb7bc12851 100644 +--- a/sound/soc/amd/yc/acp6x-mach.c ++++ b/sound/soc/amd/yc/acp6x-mach.c +@@ -521,6 +521,13 @@ static const struct dmi_system_id yc_acp_quirk_table[] = { + DMI_MATCH(DMI_PRODUCT_NAME, "OMEN by HP Gaming Laptop 16z-n000"), + } + }, ++ { ++ .driver_data = &acp6x_card, ++ .matches = { ++ DMI_MATCH(DMI_BOARD_VENDOR, "HP"), ++ DMI_MATCH(DMI_PRODUCT_NAME, "Victus by HP Gaming Laptop 15-fb2xxx"), ++ } ++ }, + { + .driver_data = &acp6x_card, + .matches = { +-- +2.39.5 + diff --git a/queue-6.15/ata-libata-acpi-do-not-assume-40-wire-cable-if-no-de.patch b/queue-6.15/ata-libata-acpi-do-not-assume-40-wire-cable-if-no-de.patch new file mode 100644 index 0000000000..b40b857387 --- /dev/null +++ b/queue-6.15/ata-libata-acpi-do-not-assume-40-wire-cable-if-no-de.patch @@ -0,0 +1,138 @@ +From 2f593edfa252886be1abc6504cfcfa0f32798924 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Mon, 19 May 2025 11:56:55 +0300 +Subject: ata: libata-acpi: Do not assume 40 wire cable if no devices are + enabled + +From: Tasos Sahanidis + +[ Upstream commit 33877220b8641b4cde474a4229ea92c0e3637883 ] + +On at least an ASRock 990FX Extreme 4 with a VIA VT6330, the devices +have not yet been enabled by the first time ata_acpi_cbl_80wire() is +called. This means that the ata_for_each_dev loop is never entered, +and a 40 wire cable is assumed. + +The VIA controller on this board does not report the cable in the PCI +config space, thus having to fall back to ACPI even though no SATA +bridge is present. + +The _GTM values are correctly reported by the firmware through ACPI, +which has already set up faster transfer modes, but due to the above +the controller is forced down to a maximum of UDMA/33. + +Resolve this by modifying ata_acpi_cbl_80wire() to directly return the +cable type. First, an unknown cable is assumed which preserves the mode +set by the firmware, and then on subsequent calls when the devices have +been enabled, an 80 wire cable is correctly detected. + +Since the function now directly returns the cable type, it is renamed +to ata_acpi_cbl_pata_type(). + +Signed-off-by: Tasos Sahanidis +Link: https://lore.kernel.org/r/20250519085945.1399466-1-tasos@tasossah.com +Signed-off-by: Niklas Cassel +Signed-off-by: Sasha Levin +--- + drivers/ata/libata-acpi.c | 24 ++++++++++++++++-------- + drivers/ata/pata_via.c | 6 ++---- + include/linux/libata.h | 7 +++---- + 3 files changed, 21 insertions(+), 16 deletions(-) + +diff --git a/drivers/ata/libata-acpi.c b/drivers/ata/libata-acpi.c +index b7f0bf7955213..f2140fc06ba0f 100644 +--- a/drivers/ata/libata-acpi.c ++++ b/drivers/ata/libata-acpi.c +@@ -514,15 +514,19 @@ unsigned int ata_acpi_gtm_xfermask(struct ata_device *dev, + EXPORT_SYMBOL_GPL(ata_acpi_gtm_xfermask); + + /** +- * ata_acpi_cbl_80wire - Check for 80 wire cable ++ * ata_acpi_cbl_pata_type - Return PATA cable type + * @ap: Port to check +- * @gtm: GTM data to use + * +- * Return 1 if the @gtm indicates the BIOS selected an 80wire mode. ++ * Return ATA_CBL_PATA* according to the transfer mode selected by BIOS + */ +-int ata_acpi_cbl_80wire(struct ata_port *ap, const struct ata_acpi_gtm *gtm) ++int ata_acpi_cbl_pata_type(struct ata_port *ap) + { + struct ata_device *dev; ++ int ret = ATA_CBL_PATA_UNK; ++ const struct ata_acpi_gtm *gtm = ata_acpi_init_gtm(ap); ++ ++ if (!gtm) ++ return ATA_CBL_PATA40; + + ata_for_each_dev(dev, &ap->link, ENABLED) { + unsigned int xfer_mask, udma_mask; +@@ -530,13 +534,17 @@ int ata_acpi_cbl_80wire(struct ata_port *ap, const struct ata_acpi_gtm *gtm) + xfer_mask = ata_acpi_gtm_xfermask(dev, gtm); + ata_unpack_xfermask(xfer_mask, NULL, NULL, &udma_mask); + +- if (udma_mask & ~ATA_UDMA_MASK_40C) +- return 1; ++ ret = ATA_CBL_PATA40; ++ ++ if (udma_mask & ~ATA_UDMA_MASK_40C) { ++ ret = ATA_CBL_PATA80; ++ break; ++ } + } + +- return 0; ++ return ret; + } +-EXPORT_SYMBOL_GPL(ata_acpi_cbl_80wire); ++EXPORT_SYMBOL_GPL(ata_acpi_cbl_pata_type); + + static void ata_acpi_gtf_to_tf(struct ata_device *dev, + const struct ata_acpi_gtf *gtf, +diff --git a/drivers/ata/pata_via.c b/drivers/ata/pata_via.c +index d82728a01832b..bb80e7800dcbe 100644 +--- a/drivers/ata/pata_via.c ++++ b/drivers/ata/pata_via.c +@@ -201,11 +201,9 @@ static int via_cable_detect(struct ata_port *ap) { + two drives */ + if (ata66 & (0x10100000 >> (16 * ap->port_no))) + return ATA_CBL_PATA80; ++ + /* Check with ACPI so we can spot BIOS reported SATA bridges */ +- if (ata_acpi_init_gtm(ap) && +- ata_acpi_cbl_80wire(ap, ata_acpi_init_gtm(ap))) +- return ATA_CBL_PATA80; +- return ATA_CBL_PATA40; ++ return ata_acpi_cbl_pata_type(ap); + } + + static int via_pre_reset(struct ata_link *link, unsigned long deadline) +diff --git a/include/linux/libata.h b/include/linux/libata.h +index e5695998acb02..ec3b0c9c2a8cf 100644 +--- a/include/linux/libata.h ++++ b/include/linux/libata.h +@@ -1363,7 +1363,7 @@ int ata_acpi_stm(struct ata_port *ap, const struct ata_acpi_gtm *stm); + int ata_acpi_gtm(struct ata_port *ap, struct ata_acpi_gtm *stm); + unsigned int ata_acpi_gtm_xfermask(struct ata_device *dev, + const struct ata_acpi_gtm *gtm); +-int ata_acpi_cbl_80wire(struct ata_port *ap, const struct ata_acpi_gtm *gtm); ++int ata_acpi_cbl_pata_type(struct ata_port *ap); + #else + static inline const struct ata_acpi_gtm *ata_acpi_init_gtm(struct ata_port *ap) + { +@@ -1388,10 +1388,9 @@ static inline unsigned int ata_acpi_gtm_xfermask(struct ata_device *dev, + return 0; + } + +-static inline int ata_acpi_cbl_80wire(struct ata_port *ap, +- const struct ata_acpi_gtm *gtm) ++static inline int ata_acpi_cbl_pata_type(struct ata_port *ap) + { +- return 0; ++ return ATA_CBL_PATA40; + } + #endif + +-- +2.39.5 + diff --git a/queue-6.15/ata-pata_cs5536-fix-build-on-32-bit-uml.patch b/queue-6.15/ata-pata_cs5536-fix-build-on-32-bit-uml.patch new file mode 100644 index 0000000000..d45701dd88 --- /dev/null +++ b/queue-6.15/ata-pata_cs5536-fix-build-on-32-bit-uml.patch @@ -0,0 +1,38 @@ +From 375213a74a54bf3fc19edf459da165d5055c7f80 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Fri, 6 Jun 2025 11:01:11 +0200 +Subject: ata: pata_cs5536: fix build on 32-bit UML + +From: Johannes Berg + +[ Upstream commit fe5b391fc56f77cf3c22a9dd4f0ce20db0e3533f ] + +On 32-bit ARCH=um, CONFIG_X86_32 is still defined, so it +doesn't indicate building on real X86 machines. There's +no MSR on UML though, so add a check for CONFIG_X86. + +Reported-by: Arnd Bergmann +Signed-off-by: Johannes Berg +Link: https://lore.kernel.org/r/20250606090110.15784-2-johannes@sipsolutions.net +Signed-off-by: Niklas Cassel +Signed-off-by: Sasha Levin +--- + drivers/ata/pata_cs5536.c | 2 +- + 1 file changed, 1 insertion(+), 1 deletion(-) + +diff --git a/drivers/ata/pata_cs5536.c b/drivers/ata/pata_cs5536.c +index b811efd2cc346..73e81e160c91f 100644 +--- a/drivers/ata/pata_cs5536.c ++++ b/drivers/ata/pata_cs5536.c +@@ -27,7 +27,7 @@ + #include + #include + +-#ifdef CONFIG_X86_32 ++#if defined(CONFIG_X86) && defined(CONFIG_X86_32) + #include + static int use_msr; + module_param_named(msr, use_msr, int, 0644); +-- +2.39.5 + diff --git a/queue-6.15/bluetooth-prevent-unintended-pause-by-checking-if-ad.patch b/queue-6.15/bluetooth-prevent-unintended-pause-by-checking-if-ad.patch new file mode 100644 index 0000000000..b7abc14189 --- /dev/null +++ b/queue-6.15/bluetooth-prevent-unintended-pause-by-checking-if-ad.patch @@ -0,0 +1,69 @@ +From 18b02af3501cefc9a3476cbdf15b992b4344973c Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Thu, 19 Jun 2025 11:01:07 +0800 +Subject: Bluetooth: Prevent unintended pause by checking if advertising is + active + +From: Yang Li + +[ Upstream commit 1f029b4e30a602db33dedee5ac676e9236ad193c ] + +When PA Create Sync is enabled, advertising resumes unexpectedly. +Therefore, it's necessary to check whether advertising is currently +active before attempting to pause it. + + < HCI Command: LE Add Device To... (0x08|0x0011) plen 7 #1345 [hci0] 48.306205 + Address type: Random (0x01) + Address: 4F:84:84:5F:88:17 (Resolvable) + Identity type: Random (0x01) + Identity: FC:5B:8C:F7:5D:FB (Static) + < HCI Command: LE Set Address Re.. (0x08|0x002d) plen 1 #1347 [hci0] 48.308023 + Address resolution: Enabled (0x01) + ... + < HCI Command: LE Set Extended A.. (0x08|0x0039) plen 6 #1349 [hci0] 48.309650 + Extended advertising: Enabled (0x01) + Number of sets: 1 (0x01) + Entry 0 + Handle: 0x01 + Duration: 0 ms (0x00) + Max ext adv events: 0 + ... + < HCI Command: LE Periodic Adve.. (0x08|0x0044) plen 14 #1355 [hci0] 48.314575 + Options: 0x0000 + Use advertising SID, Advertiser Address Type and address + Reporting initially enabled + SID: 0x02 + Adv address type: Random (0x01) + Adv address: 4F:84:84:5F:88:17 (Resolvable) + Identity type: Random (0x01) + Identity: FC:5B:8C:F7:5D:FB (Static) + Skip: 0x0000 + Sync timeout: 20000 msec (0x07d0) + Sync CTE type: 0x0000 + +Fixes: ad383c2c65a5 ("Bluetooth: hci_sync: Enable advertising when LL privacy is enabled") +Signed-off-by: Yang Li +Signed-off-by: Luiz Augusto von Dentz +Signed-off-by: Sasha Levin +--- + net/bluetooth/hci_sync.c | 4 ++++ + 1 file changed, 4 insertions(+) + +diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c +index 3660da3b241a7..9955d6cd7b76f 100644 +--- a/net/bluetooth/hci_sync.c ++++ b/net/bluetooth/hci_sync.c +@@ -2507,6 +2507,10 @@ static int hci_pause_advertising_sync(struct hci_dev *hdev) + int err; + int old_state; + ++ /* If controller is not advertising we are done. */ ++ if (!hci_dev_test_flag(hdev, HCI_LE_ADV)) ++ return 0; ++ + /* If already been paused there is nothing to do. */ + if (hdev->advertising_paused) + return 0; +-- +2.39.5 + diff --git a/queue-6.15/btrfs-fix-failure-to-rebuild-free-space-tree-using-m.patch b/queue-6.15/btrfs-fix-failure-to-rebuild-free-space-tree-using-m.patch new file mode 100644 index 0000000000..8c43708ca6 --- /dev/null +++ b/queue-6.15/btrfs-fix-failure-to-rebuild-free-space-tree-using-m.patch @@ -0,0 +1,190 @@ +From a928455718bd18356f3c58407fa23d51c0b69c52 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Sat, 7 Jun 2025 13:43:32 +0100 +Subject: btrfs: fix failure to rebuild free space tree using multiple + transactions + +From: Filipe Manana + +[ Upstream commit 1e6ed33cabba8f06f532f2e5851a102602823734 ] + +If we are rebuilding a free space tree, while modifying the free space +tree we may need to allocate a new metadata block group. +If we end up using multiple transactions for the rebuild, when we call +btrfs_end_transaction() we enter btrfs_create_pending_block_groups() +which calls add_block_group_free_space() to add items to the free space +tree for the block group. + +Then later during the free space tree rebuild, at +btrfs_rebuild_free_space_tree(), we may find such new block groups +and call populate_free_space_tree() for them, which fails with -EEXIST +because there are already items in the free space tree. Then we abort the +transaction with -EEXIST at btrfs_rebuild_free_space_tree(). +Notice that we say "may find" the new block groups because a new block +group may be inserted in the block groups rbtree, which is being iterated +by the rebuild process, before or after the current node where the rebuild +process is currently at. + +Syzbot recently reported such case which produces a trace like the +following: + + ------------[ cut here ]------------ + BTRFS: Transaction aborted (error -17) + WARNING: CPU: 1 PID: 7626 at fs/btrfs/free-space-tree.c:1341 btrfs_rebuild_free_space_tree+0x470/0x54c fs/btrfs/free-space-tree.c:1341 + Modules linked in: + CPU: 1 UID: 0 PID: 7626 Comm: syz.2.25 Not tainted 6.15.0-rc7-syzkaller-00085-gd7fa1af5b33e-dirty #0 PREEMPT + Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 05/07/2025 + pstate: 60400005 (nZCv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--) + pc : btrfs_rebuild_free_space_tree+0x470/0x54c fs/btrfs/free-space-tree.c:1341 + lr : btrfs_rebuild_free_space_tree+0x470/0x54c fs/btrfs/free-space-tree.c:1341 + sp : ffff80009c4f7740 + x29: ffff80009c4f77b0 x28: ffff0000d4c3f400 x27: 0000000000000000 + x26: dfff800000000000 x25: ffff70001389eee8 x24: 0000000000000003 + x23: 1fffe000182b6e7b x22: 0000000000000000 x21: ffff0000c15b73d8 + x20: 00000000ffffffef x19: ffff0000c15b7378 x18: 1fffe0003386f276 + x17: ffff80008f31e000 x16: ffff80008adbe98c x15: 0000000000000001 + x14: 1fffe0001b281550 x13: 0000000000000000 x12: 0000000000000000 + x11: ffff60001b281551 x10: 0000000000000003 x9 : 1c8922000a902c00 + x8 : 1c8922000a902c00 x7 : ffff800080485878 x6 : 0000000000000000 + x5 : 0000000000000001 x4 : 0000000000000001 x3 : ffff80008047843c + x2 : 0000000000000001 x1 : ffff80008b3ebc40 x0 : 0000000000000001 + Call trace: + btrfs_rebuild_free_space_tree+0x470/0x54c fs/btrfs/free-space-tree.c:1341 (P) + btrfs_start_pre_rw_mount+0xa78/0xe10 fs/btrfs/disk-io.c:3074 + btrfs_remount_rw fs/btrfs/super.c:1319 [inline] + btrfs_reconfigure+0x828/0x2418 fs/btrfs/super.c:1543 + reconfigure_super+0x1d4/0x6f0 fs/super.c:1083 + do_remount fs/namespace.c:3365 [inline] + path_mount+0xb34/0xde0 fs/namespace.c:4200 + do_mount fs/namespace.c:4221 [inline] + __do_sys_mount fs/namespace.c:4432 [inline] + __se_sys_mount fs/namespace.c:4409 [inline] + __arm64_sys_mount+0x3e8/0x468 fs/namespace.c:4409 + __invoke_syscall arch/arm64/kernel/syscall.c:35 [inline] + invoke_syscall+0x98/0x2b8 arch/arm64/kernel/syscall.c:49 + el0_svc_common+0x130/0x23c arch/arm64/kernel/syscall.c:132 + do_el0_svc+0x48/0x58 arch/arm64/kernel/syscall.c:151 + el0_svc+0x58/0x17c arch/arm64/kernel/entry-common.c:767 + el0t_64_sync_handler+0x78/0x108 arch/arm64/kernel/entry-common.c:786 + el0t_64_sync+0x198/0x19c arch/arm64/kernel/entry.S:600 + irq event stamp: 330 + hardirqs last enabled at (329): [] raw_spin_rq_unlock_irq kernel/sched/sched.h:1525 [inline] + hardirqs last enabled at (329): [] finish_lock_switch+0xb0/0x1c0 kernel/sched/core.c:5130 + hardirqs last disabled at (330): [] el1_dbg+0x24/0x80 arch/arm64/kernel/entry-common.c:511 + softirqs last enabled at (10): [] local_bh_enable+0x10/0x34 include/linux/bottom_half.h:32 + softirqs last disabled at (8): [] local_bh_disable+0x10/0x34 include/linux/bottom_half.h:19 + ---[ end trace 0000000000000000 ]--- + +Fix this by flagging new block groups which had their free space tree +entries already added and then skip them in the rebuild process. Also, +since the rebuild may be triggered when doing a remount, make sure that +when we clear an existing free space tree that we clear such flag from +every existing block group, otherwise we would skip those block groups +during the rebuild. + +Reported-by: syzbot+d0014fb0fc39c5487ae5@syzkaller.appspotmail.com +Link: https://lore.kernel.org/linux-btrfs/68460a54.050a0220.daf97.0af5.GAE@google.com/ +Fixes: 882af9f13e83 ("btrfs: handle free space tree rebuild in multiple transactions") +Reviewed-by: Boris Burkov +Signed-off-by: Filipe Manana +Signed-off-by: David Sterba +Signed-off-by: Sasha Levin +--- + fs/btrfs/block-group.h | 2 ++ + fs/btrfs/free-space-tree.c | 40 ++++++++++++++++++++++++++++++++++++++ + 2 files changed, 42 insertions(+) + +diff --git a/fs/btrfs/block-group.h b/fs/btrfs/block-group.h +index 36937eeab9b8a..08548b0b5ea1d 100644 +--- a/fs/btrfs/block-group.h ++++ b/fs/btrfs/block-group.h +@@ -83,6 +83,8 @@ enum btrfs_block_group_flags { + BLOCK_GROUP_FLAG_ZONED_DATA_RELOC, + /* Does the block group need to be added to the free space tree? */ + BLOCK_GROUP_FLAG_NEEDS_FREE_SPACE, ++ /* Set after we add a new block group to the free space tree. */ ++ BLOCK_GROUP_FLAG_FREE_SPACE_ADDED, + /* Indicate that the block group is placed on a sequential zone */ + BLOCK_GROUP_FLAG_SEQUENTIAL_ZONE, + /* +diff --git a/fs/btrfs/free-space-tree.c b/fs/btrfs/free-space-tree.c +index 39c6b96a4c25a..b65a20fd519ba 100644 +--- a/fs/btrfs/free-space-tree.c ++++ b/fs/btrfs/free-space-tree.c +@@ -1217,6 +1217,7 @@ static int clear_free_space_tree(struct btrfs_trans_handle *trans, + { + BTRFS_PATH_AUTO_FREE(path); + struct btrfs_key key; ++ struct rb_node *node; + int nr; + int ret; + +@@ -1245,6 +1246,16 @@ static int clear_free_space_tree(struct btrfs_trans_handle *trans, + btrfs_release_path(path); + } + ++ node = rb_first_cached(&trans->fs_info->block_group_cache_tree); ++ while (node) { ++ struct btrfs_block_group *bg; ++ ++ bg = rb_entry(node, struct btrfs_block_group, cache_node); ++ clear_bit(BLOCK_GROUP_FLAG_FREE_SPACE_ADDED, &bg->runtime_flags); ++ node = rb_next(node); ++ cond_resched(); ++ } ++ + return 0; + } + +@@ -1334,12 +1345,18 @@ int btrfs_rebuild_free_space_tree(struct btrfs_fs_info *fs_info) + + block_group = rb_entry(node, struct btrfs_block_group, + cache_node); ++ ++ if (test_bit(BLOCK_GROUP_FLAG_FREE_SPACE_ADDED, ++ &block_group->runtime_flags)) ++ goto next; ++ + ret = populate_free_space_tree(trans, block_group); + if (ret) { + btrfs_abort_transaction(trans, ret); + btrfs_end_transaction(trans); + return ret; + } ++next: + if (btrfs_should_end_transaction(trans)) { + btrfs_end_transaction(trans); + trans = btrfs_start_transaction(free_space_root, 1); +@@ -1366,6 +1383,29 @@ static int __add_block_group_free_space(struct btrfs_trans_handle *trans, + + clear_bit(BLOCK_GROUP_FLAG_NEEDS_FREE_SPACE, &block_group->runtime_flags); + ++ /* ++ * While rebuilding the free space tree we may allocate new metadata ++ * block groups while modifying the free space tree. ++ * ++ * Because during the rebuild (at btrfs_rebuild_free_space_tree()) we ++ * can use multiple transactions, every time btrfs_end_transaction() is ++ * called at btrfs_rebuild_free_space_tree() we finish the creation of ++ * new block groups by calling btrfs_create_pending_block_groups(), and ++ * that in turn calls us, through add_block_group_free_space(), to add ++ * a free space info item and a free space extent item for the block ++ * group. ++ * ++ * Then later btrfs_rebuild_free_space_tree() may find such new block ++ * groups and processes them with populate_free_space_tree(), which can ++ * fail with EEXIST since there are already items for the block group in ++ * the free space tree. Notice that we say "may find" because a new ++ * block group may be added to the block groups rbtree in a node before ++ * or after the block group currently being processed by the rebuild ++ * process. So signal the rebuild process to skip such new block groups ++ * if it finds them. ++ */ ++ set_bit(BLOCK_GROUP_FLAG_FREE_SPACE_ADDED, &block_group->runtime_flags); ++ + ret = add_new_free_space_info(trans, block_group, path); + if (ret) + return ret; +-- +2.39.5 + diff --git a/queue-6.15/btrfs-fix-inode-lookup-error-handling-during-log-rep.patch b/queue-6.15/btrfs-fix-inode-lookup-error-handling-during-log-rep.patch new file mode 100644 index 0000000000..ec26ef0f32 --- /dev/null +++ b/queue-6.15/btrfs-fix-inode-lookup-error-handling-during-log-rep.patch @@ -0,0 +1,295 @@ +From bd9bca9af82723d88a926f90770ec16e840d68e5 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Wed, 18 Jun 2025 15:58:31 +0100 +Subject: btrfs: fix inode lookup error handling during log replay + +From: Filipe Manana + +[ Upstream commit 5f61b961599acbd2bed028d3089105a1f7d224b8 ] + +When replaying log trees we use read_one_inode() to get an inode, which is +just a wrapper around btrfs_iget_logging(), which in turn is a wrapper for +btrfs_iget(). But read_one_inode() always returns NULL for any error +that btrfs_iget_logging() / btrfs_iget() may return and this is a problem +because: + +1) In many callers of read_one_inode() we convert the NULL into -EIO, + which is not accurate since btrfs_iget() may return -ENOMEM and -ENOENT + for example, besides -EIO and other errors. So during log replay we + may end up reporting a false -EIO, which is confusing since we may + not have had any IO error at all; + +2) When replaying directory deletes, at replay_dir_deletes(), we assume + the NULL returned from read_one_inode() means that the inode doesn't + exist and then proceed as if no error had happened. This is wrong + because unless btrfs_iget() returned ERR_PTR(-ENOENT), we had an + actual error and the target inode may exist in the target subvolume + root - this may later result in the log replay code failing at a + later stage (if we are "lucky") or succeed but leaving some + inconsistency in the filesystem. + +So fix this by not ignoring errors from btrfs_iget_logging() and as +a consequence remove the read_one_inode() wrapper and just use +btrfs_iget_logging() directly. Also since btrfs_iget_logging() is +supposed to be called only against subvolume roots, just like +read_one_inode() which had a comment about it, add an assertion to +btrfs_iget_logging() to check that the target root corresponds to a +subvolume root. + +Fixes: 5d4f98a28c7d ("Btrfs: Mixed back reference (FORWARD ROLLING FORMAT CHANGE)") +Reviewed-by: Johannes Thumshirn +Reviewed-by: Qu Wenruo +Signed-off-by: Filipe Manana +Signed-off-by: David Sterba +Signed-off-by: Sasha Levin +--- + fs/btrfs/tree-log.c | 127 +++++++++++++++++++++----------------------- + 1 file changed, 62 insertions(+), 65 deletions(-) + +diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c +index 118db10464207..f8611ec11da9d 100644 +--- a/fs/btrfs/tree-log.c ++++ b/fs/btrfs/tree-log.c +@@ -143,6 +143,9 @@ static struct btrfs_inode *btrfs_iget_logging(u64 objectid, struct btrfs_root *r + unsigned int nofs_flag; + struct btrfs_inode *inode; + ++ /* Only meant to be called for subvolume roots and not for log roots. */ ++ ASSERT(is_fstree(btrfs_root_id(root))); ++ + /* + * We're holding a transaction handle whether we are logging or + * replaying a log tree, so we must make sure NOFS semantics apply +@@ -604,21 +607,6 @@ static int read_alloc_one_name(struct extent_buffer *eb, void *start, int len, + return 0; + } + +-/* +- * simple helper to read an inode off the disk from a given root +- * This can only be called for subvolume roots and not for the log +- */ +-static noinline struct btrfs_inode *read_one_inode(struct btrfs_root *root, +- u64 objectid) +-{ +- struct btrfs_inode *inode; +- +- inode = btrfs_iget_logging(objectid, root); +- if (IS_ERR(inode)) +- return NULL; +- return inode; +-} +- + /* replays a single extent in 'eb' at 'slot' with 'key' into the + * subvolume 'root'. path is released on entry and should be released + * on exit. +@@ -671,9 +659,9 @@ static noinline int replay_one_extent(struct btrfs_trans_handle *trans, + return 0; + } + +- inode = read_one_inode(root, key->objectid); +- if (!inode) +- return -EIO; ++ inode = btrfs_iget_logging(key->objectid, root); ++ if (IS_ERR(inode)) ++ return PTR_ERR(inode); + + /* + * first check to see if we already have this extent in the +@@ -945,9 +933,10 @@ static noinline int drop_one_dir_item(struct btrfs_trans_handle *trans, + + btrfs_release_path(path); + +- inode = read_one_inode(root, location.objectid); +- if (!inode) { +- ret = -EIO; ++ inode = btrfs_iget_logging(location.objectid, root); ++ if (IS_ERR(inode)) { ++ ret = PTR_ERR(inode); ++ inode = NULL; + goto out; + } + +@@ -1166,10 +1155,10 @@ static inline int __add_inode_ref(struct btrfs_trans_handle *trans, + kfree(victim_name.name); + return ret; + } else if (!ret) { +- ret = -ENOENT; +- victim_parent = read_one_inode(root, +- parent_objectid); +- if (victim_parent) { ++ victim_parent = btrfs_iget_logging(parent_objectid, root); ++ if (IS_ERR(victim_parent)) { ++ ret = PTR_ERR(victim_parent); ++ } else { + inc_nlink(&inode->vfs_inode); + btrfs_release_path(path); + +@@ -1314,9 +1303,9 @@ static int unlink_old_inode_refs(struct btrfs_trans_handle *trans, + struct btrfs_inode *dir; + + btrfs_release_path(path); +- dir = read_one_inode(root, parent_id); +- if (!dir) { +- ret = -ENOENT; ++ dir = btrfs_iget_logging(parent_id, root); ++ if (IS_ERR(dir)) { ++ ret = PTR_ERR(dir); + kfree(name.name); + goto out; + } +@@ -1388,15 +1377,17 @@ static noinline int add_inode_ref(struct btrfs_trans_handle *trans, + * copy the back ref in. The link count fixup code will take + * care of the rest + */ +- dir = read_one_inode(root, parent_objectid); +- if (!dir) { +- ret = -ENOENT; ++ dir = btrfs_iget_logging(parent_objectid, root); ++ if (IS_ERR(dir)) { ++ ret = PTR_ERR(dir); ++ dir = NULL; + goto out; + } + +- inode = read_one_inode(root, inode_objectid); +- if (!inode) { +- ret = -EIO; ++ inode = btrfs_iget_logging(inode_objectid, root); ++ if (IS_ERR(inode)) { ++ ret = PTR_ERR(inode); ++ inode = NULL; + goto out; + } + +@@ -1408,11 +1399,13 @@ static noinline int add_inode_ref(struct btrfs_trans_handle *trans, + * parent object can change from one array + * item to another. + */ +- if (!dir) +- dir = read_one_inode(root, parent_objectid); + if (!dir) { +- ret = -ENOENT; +- goto out; ++ dir = btrfs_iget_logging(parent_objectid, root); ++ if (IS_ERR(dir)) { ++ ret = PTR_ERR(dir); ++ dir = NULL; ++ goto out; ++ } + } + } else { + ret = ref_get_fields(eb, ref_ptr, &name, &ref_index); +@@ -1681,9 +1674,9 @@ static noinline int fixup_inode_link_counts(struct btrfs_trans_handle *trans, + break; + + btrfs_release_path(path); +- inode = read_one_inode(root, key.offset); +- if (!inode) { +- ret = -EIO; ++ inode = btrfs_iget_logging(key.offset, root); ++ if (IS_ERR(inode)) { ++ ret = PTR_ERR(inode); + break; + } + +@@ -1719,9 +1712,9 @@ static noinline int link_to_fixup_dir(struct btrfs_trans_handle *trans, + struct btrfs_inode *inode; + struct inode *vfs_inode; + +- inode = read_one_inode(root, objectid); +- if (!inode) +- return -EIO; ++ inode = btrfs_iget_logging(objectid, root); ++ if (IS_ERR(inode)) ++ return PTR_ERR(inode); + + vfs_inode = &inode->vfs_inode; + key.objectid = BTRFS_TREE_LOG_FIXUP_OBJECTID; +@@ -1760,14 +1753,14 @@ static noinline int insert_one_name(struct btrfs_trans_handle *trans, + struct btrfs_inode *dir; + int ret; + +- inode = read_one_inode(root, location->objectid); +- if (!inode) +- return -ENOENT; ++ inode = btrfs_iget_logging(location->objectid, root); ++ if (IS_ERR(inode)) ++ return PTR_ERR(inode); + +- dir = read_one_inode(root, dirid); +- if (!dir) { ++ dir = btrfs_iget_logging(dirid, root); ++ if (IS_ERR(dir)) { + iput(&inode->vfs_inode); +- return -EIO; ++ return PTR_ERR(dir); + } + + ret = btrfs_add_link(trans, dir, inode, name, 1, index); +@@ -1844,9 +1837,9 @@ static noinline int replay_one_name(struct btrfs_trans_handle *trans, + bool update_size = true; + bool name_added = false; + +- dir = read_one_inode(root, key->objectid); +- if (!dir) +- return -EIO; ++ dir = btrfs_iget_logging(key->objectid, root); ++ if (IS_ERR(dir)) ++ return PTR_ERR(dir); + + ret = read_alloc_one_name(eb, di + 1, btrfs_dir_name_len(eb, di), &name); + if (ret) +@@ -2146,9 +2139,10 @@ static noinline int check_item_in_log(struct btrfs_trans_handle *trans, + btrfs_dir_item_key_to_cpu(eb, di, &location); + btrfs_release_path(path); + btrfs_release_path(log_path); +- inode = read_one_inode(root, location.objectid); +- if (!inode) { +- ret = -EIO; ++ inode = btrfs_iget_logging(location.objectid, root); ++ if (IS_ERR(inode)) { ++ ret = PTR_ERR(inode); ++ inode = NULL; + goto out; + } + +@@ -2300,14 +2294,17 @@ static noinline int replay_dir_deletes(struct btrfs_trans_handle *trans, + if (!log_path) + return -ENOMEM; + +- dir = read_one_inode(root, dirid); +- /* it isn't an error if the inode isn't there, that can happen +- * because we replay the deletes before we copy in the inode item +- * from the log ++ dir = btrfs_iget_logging(dirid, root); ++ /* ++ * It isn't an error if the inode isn't there, that can happen because ++ * we replay the deletes before we copy in the inode item from the log. + */ +- if (!dir) { ++ if (IS_ERR(dir)) { + btrfs_free_path(log_path); +- return 0; ++ ret = PTR_ERR(dir); ++ if (ret == -ENOENT) ++ ret = 0; ++ return ret; + } + + range_start = 0; +@@ -2466,9 +2463,9 @@ static int replay_one_buffer(struct btrfs_root *log, struct extent_buffer *eb, + struct btrfs_inode *inode; + u64 from; + +- inode = read_one_inode(root, key.objectid); +- if (!inode) { +- ret = -EIO; ++ inode = btrfs_iget_logging(key.objectid, root); ++ if (IS_ERR(inode)) { ++ ret = PTR_ERR(inode); + break; + } + from = ALIGN(i_size_read(&inode->vfs_inode), +-- +2.39.5 + diff --git a/queue-6.15/btrfs-fix-iteration-of-extrefs-during-log-replay.patch b/queue-6.15/btrfs-fix-iteration-of-extrefs-during-log-replay.patch new file mode 100644 index 0000000000..a39ea2bab6 --- /dev/null +++ b/queue-6.15/btrfs-fix-iteration-of-extrefs-during-log-replay.patch @@ -0,0 +1,51 @@ +From 9f3597b6313a535a6c6d70b27db36d524dfabb40 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Mon, 23 Jun 2025 12:11:58 +0100 +Subject: btrfs: fix iteration of extrefs during log replay + +From: Filipe Manana + +[ Upstream commit 54a7081ed168b72a8a2d6ef4ba3a1259705a2926 ] + +At __inode_add_ref() when processing extrefs, if we jump into the next +label we have an undefined value of victim_name.len, since we haven't +initialized it before we did the goto. This results in an invalid memory +access in the next iteration of the loop since victim_name.len was not +initialized to the length of the name of the current extref. + +Fix this by initializing victim_name.len with the current extref's name +length. + +Fixes: e43eec81c516 ("btrfs: use struct qstr instead of name and namelen pairs") +Reviewed-by: Johannes Thumshirn +Reviewed-by: Qu Wenruo +Signed-off-by: Filipe Manana +Signed-off-by: David Sterba +Signed-off-by: Sasha Levin +--- + fs/btrfs/tree-log.c | 4 ++-- + 1 file changed, 2 insertions(+), 2 deletions(-) + +diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c +index cc223c3b39c10..118db10464207 100644 +--- a/fs/btrfs/tree-log.c ++++ b/fs/btrfs/tree-log.c +@@ -1145,13 +1145,13 @@ static inline int __add_inode_ref(struct btrfs_trans_handle *trans, + struct fscrypt_str victim_name; + + extref = (struct btrfs_inode_extref *)(base + cur_offset); ++ victim_name.len = btrfs_inode_extref_name_len(leaf, extref); + + if (btrfs_inode_extref_parent(leaf, extref) != parent_objectid) + goto next; + + ret = read_alloc_one_name(leaf, &extref->name, +- btrfs_inode_extref_name_len(leaf, extref), +- &victim_name); ++ victim_name.len, &victim_name); + if (ret) + return ret; + +-- +2.39.5 + diff --git a/queue-6.15/btrfs-fix-missing-error-handling-when-searching-for-.patch b/queue-6.15/btrfs-fix-missing-error-handling-when-searching-for-.patch new file mode 100644 index 0000000000..ba9e98bb5d --- /dev/null +++ b/queue-6.15/btrfs-fix-missing-error-handling-when-searching-for-.patch @@ -0,0 +1,45 @@ +From 36c48910bcf47e502c02bd143177254338ad5b55 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Wed, 18 Jun 2025 16:57:07 +0100 +Subject: btrfs: fix missing error handling when searching for inode refs + during log replay + +From: Filipe Manana + +[ Upstream commit 6561a40ceced9082f50c374a22d5966cf9fc5f5c ] + +During log replay, at __add_inode_ref(), when we are searching for inode +ref keys we totally ignore if btrfs_search_slot() returns an error. This +may make a log replay succeed when there was an actual error and leave +some metadata inconsistency in a subvolume tree. Fix this by checking if +an error was returned from btrfs_search_slot() and if so, return it to +the caller. + +Fixes: e02119d5a7b4 ("Btrfs: Add a write ahead tree log to optimize synchronous operations") +Reviewed-by: Johannes Thumshirn +Reviewed-by: Qu Wenruo +Signed-off-by: Filipe Manana +Signed-off-by: David Sterba +Signed-off-by: Sasha Levin +--- + fs/btrfs/tree-log.c | 4 +++- + 1 file changed, 3 insertions(+), 1 deletion(-) + +diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c +index ef9660eabf0c6..cc223c3b39c10 100644 +--- a/fs/btrfs/tree-log.c ++++ b/fs/btrfs/tree-log.c +@@ -1070,7 +1070,9 @@ static inline int __add_inode_ref(struct btrfs_trans_handle *trans, + search_key.type = BTRFS_INODE_REF_KEY; + search_key.offset = parent_objectid; + ret = btrfs_search_slot(NULL, root, &search_key, path, 0, 0); +- if (ret == 0) { ++ if (ret < 0) { ++ return ret; ++ } else if (ret == 0) { + struct btrfs_inode_ref *victim_ref; + unsigned long ptr; + unsigned long ptr_end; +-- +2.39.5 + diff --git a/queue-6.15/btrfs-propagate-last_unlink_trans-earlier-when-doing.patch b/queue-6.15/btrfs-propagate-last_unlink_trans-earlier-when-doing.patch new file mode 100644 index 0000000000..4ce9d31765 --- /dev/null +++ b/queue-6.15/btrfs-propagate-last_unlink_trans-earlier-when-doing.patch @@ -0,0 +1,100 @@ +From b4765dc4d0808eb551f517604f5cdff6738ea9a5 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Fri, 20 Jun 2025 15:54:05 +0100 +Subject: btrfs: propagate last_unlink_trans earlier when doing a rmdir + +From: Filipe Manana + +[ Upstream commit c466e33e729a0ee017d10d919cba18f503853c60 ] + +In case the removed directory had a snapshot that was deleted, we are +propagating its inode's last_unlink_trans to the parent directory after +we removed the entry from the parent directory. This leaves a small race +window where someone can log the parent directory after we removed the +entry and before we updated last_unlink_trans, and as a result if we ever +try to replay such a log tree, we will fail since we will attempt to +remove a snapshot during log replay, which is currently not possible and +results in the log replay (and mount) to fail. This is the type of failure +described in commit 1ec9a1ae1e30 ("Btrfs: fix unreplayable log after +snapshot delete + parent dir fsync"). + +So fix this by propagating the last_unlink_trans to the parent directory +before we remove the entry from it. + +Fixes: 44f714dae50a ("Btrfs: improve performance on fsync against new inode after rename/unlink") +Reviewed-by: Johannes Thumshirn +Signed-off-by: Filipe Manana +Signed-off-by: David Sterba +Signed-off-by: Sasha Levin +--- + fs/btrfs/inode.c | 36 ++++++++++++++++++------------------ + 1 file changed, 18 insertions(+), 18 deletions(-) + +diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c +index 391172b443e50..24db00a9319b3 100644 +--- a/fs/btrfs/inode.c ++++ b/fs/btrfs/inode.c +@@ -4724,7 +4724,6 @@ static int btrfs_rmdir(struct inode *dir, struct dentry *dentry) + struct btrfs_fs_info *fs_info = BTRFS_I(inode)->root->fs_info; + int ret = 0; + struct btrfs_trans_handle *trans; +- u64 last_unlink_trans; + struct fscrypt_name fname; + + if (inode->i_size > BTRFS_EMPTY_DIR_SIZE) +@@ -4750,6 +4749,23 @@ static int btrfs_rmdir(struct inode *dir, struct dentry *dentry) + goto out_notrans; + } + ++ /* ++ * Propagate the last_unlink_trans value of the deleted dir to its ++ * parent directory. This is to prevent an unrecoverable log tree in the ++ * case we do something like this: ++ * 1) create dir foo ++ * 2) create snapshot under dir foo ++ * 3) delete the snapshot ++ * 4) rmdir foo ++ * 5) mkdir foo ++ * 6) fsync foo or some file inside foo ++ * ++ * This is because we can't unlink other roots when replaying the dir ++ * deletes for directory foo. ++ */ ++ if (BTRFS_I(inode)->last_unlink_trans >= trans->transid) ++ BTRFS_I(dir)->last_unlink_trans = BTRFS_I(inode)->last_unlink_trans; ++ + if (unlikely(btrfs_ino(BTRFS_I(inode)) == BTRFS_EMPTY_SUBVOL_DIR_OBJECTID)) { + ret = btrfs_unlink_subvol(trans, BTRFS_I(dir), dentry); + goto out; +@@ -4759,27 +4775,11 @@ static int btrfs_rmdir(struct inode *dir, struct dentry *dentry) + if (ret) + goto out; + +- last_unlink_trans = BTRFS_I(inode)->last_unlink_trans; +- + /* now the directory is empty */ + ret = btrfs_unlink_inode(trans, BTRFS_I(dir), BTRFS_I(d_inode(dentry)), + &fname.disk_name); +- if (!ret) { ++ if (!ret) + btrfs_i_size_write(BTRFS_I(inode), 0); +- /* +- * Propagate the last_unlink_trans value of the deleted dir to +- * its parent directory. This is to prevent an unrecoverable +- * log tree in the case we do something like this: +- * 1) create dir foo +- * 2) create snapshot under dir foo +- * 3) delete the snapshot +- * 4) rmdir foo +- * 5) mkdir foo +- * 6) fsync foo or some file inside foo +- */ +- if (last_unlink_trans >= trans->transid) +- BTRFS_I(dir)->last_unlink_trans = last_unlink_trans; +- } + out: + btrfs_end_transaction(trans); + out_notrans: +-- +2.39.5 + diff --git a/queue-6.15/btrfs-record-new-subvolume-in-parent-dir-earlier-to-.patch b/queue-6.15/btrfs-record-new-subvolume-in-parent-dir-earlier-to-.patch new file mode 100644 index 0000000000..c35e53573e --- /dev/null +++ b/queue-6.15/btrfs-record-new-subvolume-in-parent-dir-earlier-to-.patch @@ -0,0 +1,70 @@ +From 84213d552a51c59cfc55732e455b3c65c96412fd Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Thu, 19 Jun 2025 13:13:38 +0100 +Subject: btrfs: record new subvolume in parent dir earlier to avoid dir + logging races + +From: Filipe Manana + +[ Upstream commit bf5bcf9a6fa070ec8a725b08db63fb1318f77366 ] + +Instead of recording that a new subvolume was created in a directory after +we add the entry do the directory, record it before adding the entry. This +is to avoid races where after creating the entry and before recording the +new subvolume in the directory (the call to btrfs_record_new_subvolume()), +another task logs the directory, so we end up with a log tree where we +logged a directory that has an entry pointing to a root that was not yet +committed, resulting in an invalid entry if the log is persisted and +replayed later due to a power failure or crash. + +Also state this requirement in the function comment for +btrfs_record_new_subvolume(), similar to what we do for the +btrfs_record_unlink_dir() and btrfs_record_snapshot_destroy(). + +Fixes: 45c4102f0d82 ("btrfs: avoid transaction commit on any fsync after subvolume creation") +Reviewed-by: Johannes Thumshirn +Signed-off-by: Filipe Manana +Signed-off-by: David Sterba +Signed-off-by: Sasha Levin +--- + fs/btrfs/ioctl.c | 4 ++-- + fs/btrfs/tree-log.c | 2 ++ + 2 files changed, 4 insertions(+), 2 deletions(-) + +diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c +index 63aeacc549457..b70ef4455610a 100644 +--- a/fs/btrfs/ioctl.c ++++ b/fs/btrfs/ioctl.c +@@ -666,14 +666,14 @@ static noinline int create_subvol(struct mnt_idmap *idmap, + goto out; + } + ++ btrfs_record_new_subvolume(trans, BTRFS_I(dir)); ++ + ret = btrfs_create_new_inode(trans, &new_inode_args); + if (ret) { + btrfs_abort_transaction(trans, ret); + goto out; + } + +- btrfs_record_new_subvolume(trans, BTRFS_I(dir)); +- + d_instantiate_new(dentry, new_inode_args.inode); + new_inode_args.inode = NULL; + +diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c +index f8611ec11da9d..e05140ce95be9 100644 +--- a/fs/btrfs/tree-log.c ++++ b/fs/btrfs/tree-log.c +@@ -7444,6 +7444,8 @@ void btrfs_record_snapshot_destroy(struct btrfs_trans_handle *trans, + * full log sync. + * Also we don't need to worry with renames, since btrfs_rename() marks the log + * for full commit when renaming a subvolume. ++ * ++ * Must be called before creating the subvolume entry in its parent directory. + */ + void btrfs_record_new_subvolume(const struct btrfs_trans_handle *trans, + struct btrfs_inode *dir) +-- +2.39.5 + diff --git a/queue-6.15/btrfs-use-btrfs_record_snapshot_destroy-during-rmdir.patch b/queue-6.15/btrfs-use-btrfs_record_snapshot_destroy-during-rmdir.patch new file mode 100644 index 0000000000..faa82da5de --- /dev/null +++ b/queue-6.15/btrfs-use-btrfs_record_snapshot_destroy-during-rmdir.patch @@ -0,0 +1,46 @@ +From 1bf781bd5e1000ba269800ee7e370aa666c10c92 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Fri, 20 Jun 2025 16:37:01 +0100 +Subject: btrfs: use btrfs_record_snapshot_destroy() during rmdir + +From: Filipe Manana + +[ Upstream commit 157501b0469969fc1ba53add5049575aadd79d80 ] + +We are setting the parent directory's last_unlink_trans directly which +may result in a concurrent task starting to log the directory not see the +update and therefore can log the directory after we removed a child +directory which had a snapshot within instead of falling back to a +transaction commit. Replaying such a log tree would result in a mount +failure since we can't currently delete snapshots (and subvolumes) during +log replay. This is the type of failure described in commit 1ec9a1ae1e30 +("Btrfs: fix unreplayable log after snapshot delete + parent dir fsync"). + +Fix this by using btrfs_record_snapshot_destroy() which updates the +last_unlink_trans field while holding the inode's log_mutex lock. + +Fixes: 44f714dae50a ("Btrfs: improve performance on fsync against new inode after rename/unlink") +Reviewed-by: Johannes Thumshirn +Signed-off-by: Filipe Manana +Signed-off-by: David Sterba +Signed-off-by: Sasha Levin +--- + fs/btrfs/inode.c | 2 +- + 1 file changed, 1 insertion(+), 1 deletion(-) + +diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c +index 24db00a9319b3..f18f4d59d389e 100644 +--- a/fs/btrfs/inode.c ++++ b/fs/btrfs/inode.c +@@ -4764,7 +4764,7 @@ static int btrfs_rmdir(struct inode *dir, struct dentry *dentry) + * deletes for directory foo. + */ + if (BTRFS_I(inode)->last_unlink_trans >= trans->transid) +- BTRFS_I(dir)->last_unlink_trans = BTRFS_I(inode)->last_unlink_trans; ++ btrfs_record_snapshot_destroy(trans, BTRFS_I(dir)); + + if (unlikely(btrfs_ino(BTRFS_I(inode)) == BTRFS_EMPTY_SUBVOL_DIR_OBJECTID)) { + ret = btrfs_unlink_subvol(trans, BTRFS_I(dir), dentry); +-- +2.39.5 + diff --git a/queue-6.15/dpaa2-eth-fix-xdp_rxq_info-leak.patch b/queue-6.15/dpaa2-eth-fix-xdp_rxq_info-leak.patch new file mode 100644 index 0000000000..6bc00d7bfa --- /dev/null +++ b/queue-6.15/dpaa2-eth-fix-xdp_rxq_info-leak.patch @@ -0,0 +1,101 @@ +From a7bf06420e6225309c3bda8ee2c7b5c269ebb670 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Thu, 26 Jun 2025 21:30:03 +0800 +Subject: dpaa2-eth: fix xdp_rxq_info leak + +From: Fushuai Wang + +[ Upstream commit 2def09ead4ad5907988b655d1e1454003aaf8297 ] + +The driver registered xdp_rxq_info structures via xdp_rxq_info_reg() +but failed to properly unregister them in error paths and during +removal. + +Fixes: d678be1dc1ec ("dpaa2-eth: add XDP_REDIRECT support") +Signed-off-by: Fushuai Wang +Reviewed-by: Simon Horman +Reviewed-by: Ioana Ciornei +Link: https://patch.msgid.link/20250626133003.80136-1-wangfushuai@baidu.com +Signed-off-by: Jakub Kicinski +Signed-off-by: Sasha Levin +--- + .../net/ethernet/freescale/dpaa2/dpaa2-eth.c | 26 +++++++++++++++++-- + 1 file changed, 24 insertions(+), 2 deletions(-) + +diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c +index 29886a8ba73f3..efd0048acd3b2 100644 +--- a/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c ++++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c +@@ -3928,6 +3928,7 @@ static int dpaa2_eth_setup_rx_flow(struct dpaa2_eth_priv *priv, + MEM_TYPE_PAGE_ORDER0, NULL); + if (err) { + dev_err(dev, "xdp_rxq_info_reg_mem_model failed\n"); ++ xdp_rxq_info_unreg(&fq->channel->xdp_rxq); + return err; + } + +@@ -4421,17 +4422,25 @@ static int dpaa2_eth_bind_dpni(struct dpaa2_eth_priv *priv) + return -EINVAL; + } + if (err) +- return err; ++ goto out; + } + + err = dpni_get_qdid(priv->mc_io, 0, priv->mc_token, + DPNI_QUEUE_TX, &priv->tx_qdid); + if (err) { + dev_err(dev, "dpni_get_qdid() failed\n"); +- return err; ++ goto out; + } + + return 0; ++ ++out: ++ while (i--) { ++ if (priv->fq[i].type == DPAA2_RX_FQ && ++ xdp_rxq_info_is_reg(&priv->fq[i].channel->xdp_rxq)) ++ xdp_rxq_info_unreg(&priv->fq[i].channel->xdp_rxq); ++ } ++ return err; + } + + /* Allocate rings for storing incoming frame descriptors */ +@@ -4814,6 +4823,17 @@ static void dpaa2_eth_del_ch_napi(struct dpaa2_eth_priv *priv) + } + } + ++static void dpaa2_eth_free_rx_xdp_rxq(struct dpaa2_eth_priv *priv) ++{ ++ int i; ++ ++ for (i = 0; i < priv->num_fqs; i++) { ++ if (priv->fq[i].type == DPAA2_RX_FQ && ++ xdp_rxq_info_is_reg(&priv->fq[i].channel->xdp_rxq)) ++ xdp_rxq_info_unreg(&priv->fq[i].channel->xdp_rxq); ++ } ++} ++ + static int dpaa2_eth_probe(struct fsl_mc_device *dpni_dev) + { + struct device *dev; +@@ -5017,6 +5037,7 @@ static int dpaa2_eth_probe(struct fsl_mc_device *dpni_dev) + free_percpu(priv->percpu_stats); + err_alloc_percpu_stats: + dpaa2_eth_del_ch_napi(priv); ++ dpaa2_eth_free_rx_xdp_rxq(priv); + err_bind: + dpaa2_eth_free_dpbps(priv); + err_dpbp_setup: +@@ -5069,6 +5090,7 @@ static void dpaa2_eth_remove(struct fsl_mc_device *ls_dev) + free_percpu(priv->percpu_extras); + + dpaa2_eth_del_ch_napi(priv); ++ dpaa2_eth_free_rx_xdp_rxq(priv); + dpaa2_eth_free_dpbps(priv); + dpaa2_eth_free_dpio(priv); + dpaa2_eth_free_dpni(priv); +-- +2.39.5 + diff --git a/queue-6.15/drm-bridge-aux-hpd-bridge-fix-assignment-of-the-of_n.patch b/queue-6.15/drm-bridge-aux-hpd-bridge-fix-assignment-of-the-of_n.patch new file mode 100644 index 0000000000..215c30460a --- /dev/null +++ b/queue-6.15/drm-bridge-aux-hpd-bridge-fix-assignment-of-the-of_n.patch @@ -0,0 +1,49 @@ +From 6e21bc3490e20d147b8b2c563de229475cea9ca2 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Sun, 8 Jun 2025 18:52:04 +0300 +Subject: drm/bridge: aux-hpd-bridge: fix assignment of the of_node + +From: Dmitry Baryshkov + +[ Upstream commit e8537cad824065b0425fb0429e762e14a08067c2 ] + +Perform fix similar to the one in the commit 85e444a68126 ("drm/bridge: +Fix assignment of the of_node of the parent to aux bridge"). + +The assignment of the of_node to the aux HPD bridge needs to mark the +of_node as reused, otherwise driver core will attempt to bind resources +like pinctrl, which is going to fail as corresponding pins are already +marked as used by the parent device. +Fix that by using the device_set_of_node_from_dev() helper instead of +assigning it directly. + +Fixes: e560518a6c2e ("drm/bridge: implement generic DP HPD bridge") +Signed-off-by: Dmitry Baryshkov +Reviewed-by: Neil Armstrong +Signed-off-by: Neil Armstrong +Link: https://lore.kernel.org/r/20250608-fix-aud-hpd-bridge-v1-1-4641a6f8e381@oss.qualcomm.com +Signed-off-by: Sasha Levin +--- + drivers/gpu/drm/bridge/aux-hpd-bridge.c | 3 ++- + 1 file changed, 2 insertions(+), 1 deletion(-) + +diff --git a/drivers/gpu/drm/bridge/aux-hpd-bridge.c b/drivers/gpu/drm/bridge/aux-hpd-bridge.c +index 48f297c78ee67..1ec4f16b9939d 100644 +--- a/drivers/gpu/drm/bridge/aux-hpd-bridge.c ++++ b/drivers/gpu/drm/bridge/aux-hpd-bridge.c +@@ -64,10 +64,11 @@ struct auxiliary_device *devm_drm_dp_hpd_bridge_alloc(struct device *parent, str + adev->id = ret; + adev->name = "dp_hpd_bridge"; + adev->dev.parent = parent; +- adev->dev.of_node = of_node_get(parent->of_node); + adev->dev.release = drm_aux_hpd_bridge_release; + adev->dev.platform_data = of_node_get(np); + ++ device_set_of_node_from_dev(&adev->dev, parent); ++ + ret = auxiliary_device_init(adev); + if (ret) { + of_node_put(adev->dev.platform_data); +-- +2.39.5 + diff --git a/queue-6.15/drm-bridge-panel-move-prepare_prev_first-handling-to.patch b/queue-6.15/drm-bridge-panel-move-prepare_prev_first-handling-to.patch new file mode 100644 index 0000000000..907badd024 --- /dev/null +++ b/queue-6.15/drm-bridge-panel-move-prepare_prev_first-handling-to.patch @@ -0,0 +1,67 @@ +From e9aa8b54f7238cacfa3533d102d206acc80348ec Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Thu, 20 Feb 2025 17:07:26 +0200 +Subject: drm/bridge: panel: move prepare_prev_first handling to + drm_panel_bridge_add_typed + +From: Dmitry Baryshkov + +[ Upstream commit eb028cd884e1b0976ff8c5944ee6650fe3ed0a6c ] + +The commit 5ea6b1702781 ("drm/panel: Add prepare_prev_first flag to +drm_panel") and commit 0974687a19c3 ("drm/bridge: panel: Set +pre_enable_prev_first from drmm_panel_bridge_add") added handling of +panel's prepare_prev_first to devm_panel_bridge_add() and +drmm_panel_bridge_add(). However if the driver calls +drm_panel_bridge_add_typed() directly, then the flag won't be handled +and thus the drm_bridge.pre_enable_prev_first will not be set. + +Move prepare_prev_first handling to the drm_panel_bridge_add_typed() so +that there is no way to miss the flag. + +Fixes: 5ea6b1702781 ("drm/panel: Add prepare_prev_first flag to drm_panel") +Fixes: 0974687a19c3 ("drm/bridge: panel: Set pre_enable_prev_first from drmm_panel_bridge_add") +Reported-by: Svyatoslav Ryhel +Closes: https://lore.kernel.org/dri-devel/CAPVz0n3YZass3Bns1m0XrFxtAC0DKbEPiW6vXimQx97G243sXw@mail.gmail.com/ +Signed-off-by: Dmitry Baryshkov +Reviewed-by: Neil Armstrong +Signed-off-by: Neil Armstrong +Link: https://lore.kernel.org/r/20250220-panel_prev_first-v1-1-b9e787825a1a@linaro.org +Signed-off-by: Sasha Levin +--- + drivers/gpu/drm/bridge/panel.c | 5 +---- + 1 file changed, 1 insertion(+), 4 deletions(-) + +diff --git a/drivers/gpu/drm/bridge/panel.c b/drivers/gpu/drm/bridge/panel.c +index 258c85c83a283..6c58d0759f227 100644 +--- a/drivers/gpu/drm/bridge/panel.c ++++ b/drivers/gpu/drm/bridge/panel.c +@@ -298,6 +298,7 @@ struct drm_bridge *drm_panel_bridge_add_typed(struct drm_panel *panel, + panel_bridge->bridge.of_node = panel->dev->of_node; + panel_bridge->bridge.ops = DRM_BRIDGE_OP_MODES; + panel_bridge->bridge.type = connector_type; ++ panel_bridge->bridge.pre_enable_prev_first = panel->prepare_prev_first; + + drm_bridge_add(&panel_bridge->bridge); + +@@ -412,8 +413,6 @@ struct drm_bridge *devm_drm_panel_bridge_add_typed(struct device *dev, + return bridge; + } + +- bridge->pre_enable_prev_first = panel->prepare_prev_first; +- + *ptr = bridge; + devres_add(dev, ptr); + +@@ -455,8 +454,6 @@ struct drm_bridge *drmm_panel_bridge_add(struct drm_device *drm, + if (ret) + return ERR_PTR(ret); + +- bridge->pre_enable_prev_first = panel->prepare_prev_first; +- + return bridge; + } + EXPORT_SYMBOL(drmm_panel_bridge_add); +-- +2.39.5 + diff --git a/queue-6.15/drm-exynos-fimd-guard-display-clock-control-with-run.patch b/queue-6.15/drm-exynos-fimd-guard-display-clock-control-with-run.patch new file mode 100644 index 0000000000..a9bb61babf --- /dev/null +++ b/queue-6.15/drm-exynos-fimd-guard-display-clock-control-with-run.patch @@ -0,0 +1,67 @@ +From 03cccb6fa3a2e53eb603b0c21dd33a182eee2d1d Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Wed, 18 Jun 2025 14:06:26 +0200 +Subject: drm/exynos: fimd: Guard display clock control with runtime PM calls + +From: Marek Szyprowski + +[ Upstream commit 5d91394f236167ac624b823820faf4aa928b889e ] + +Commit c9b1150a68d9 ("drm/atomic-helper: Re-order bridge chain pre-enable +and post-disable") changed the call sequence to the CRTC enable/disable +and bridge pre_enable/post_disable methods, so those bridge methods are +now called when CRTC is not yet enabled. + +This causes a lockup observed on Samsung Peach-Pit/Pi Chromebooks. The +source of this lockup is a call to fimd_dp_clock_enable() function, when +FIMD device is not yet runtime resumed. It worked before the mentioned +commit only because the CRTC implemented by the FIMD driver was always +enabled what guaranteed the FIMD device to be runtime resumed. + +This patch adds runtime PM guards to the fimd_dp_clock_enable() function +to enable its proper operation also when the CRTC implemented by FIMD is +not yet enabled. + +Fixes: 196e059a8a6a ("drm/exynos: convert clock_enable crtc callback to pipeline clock") +Signed-off-by: Marek Szyprowski +Reviewed-by: Tomi Valkeinen +Signed-off-by: Inki Dae +Signed-off-by: Sasha Levin +--- + drivers/gpu/drm/exynos/exynos_drm_fimd.c | 12 ++++++++++++ + 1 file changed, 12 insertions(+) + +diff --git a/drivers/gpu/drm/exynos/exynos_drm_fimd.c b/drivers/gpu/drm/exynos/exynos_drm_fimd.c +index c394cc702d7d4..205c238cc73a6 100644 +--- a/drivers/gpu/drm/exynos/exynos_drm_fimd.c ++++ b/drivers/gpu/drm/exynos/exynos_drm_fimd.c +@@ -187,6 +187,7 @@ struct fimd_context { + u32 i80ifcon; + bool i80_if; + bool suspended; ++ bool dp_clk_enabled; + wait_queue_head_t wait_vsync_queue; + atomic_t wait_vsync_event; + atomic_t win_updated; +@@ -1047,7 +1048,18 @@ static void fimd_dp_clock_enable(struct exynos_drm_clk *clk, bool enable) + struct fimd_context *ctx = container_of(clk, struct fimd_context, + dp_clk); + u32 val = enable ? DP_MIE_CLK_DP_ENABLE : DP_MIE_CLK_DISABLE; ++ ++ if (enable == ctx->dp_clk_enabled) ++ return; ++ ++ if (enable) ++ pm_runtime_resume_and_get(ctx->dev); ++ ++ ctx->dp_clk_enabled = enable; + writel(val, ctx->regs + DP_MIE_CLKCON); ++ ++ if (!enable) ++ pm_runtime_put(ctx->dev); + } + + static const struct exynos_drm_crtc_ops fimd_crtc_ops = { +-- +2.39.5 + diff --git a/queue-6.15/drm-i915-gsc-mei-interrupt-top-half-should-be-in-irq.patch b/queue-6.15/drm-i915-gsc-mei-interrupt-top-half-should-be-in-irq.patch new file mode 100644 index 0000000000..dff70b8ade --- /dev/null +++ b/queue-6.15/drm-i915-gsc-mei-interrupt-top-half-should-be-in-irq.patch @@ -0,0 +1,52 @@ +From 28a5d8899f8c150852b2adf7b8ea2211cc35ecb8 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Fri, 25 Apr 2025 23:11:07 +0800 +Subject: drm/i915/gsc: mei interrupt top half should be in irq disabled + context + +From: Junxiao Chang + +[ Upstream commit 8cadce97bf264ed478669c6f32d5603b34608335 ] + +MEI GSC interrupt comes from i915. It has top half and bottom half. +Top half is called from i915 interrupt handler. It should be in +irq disabled context. + +With RT kernel, by default i915 IRQ handler is in threaded IRQ. MEI GSC +top half might be in threaded IRQ context. generic_handle_irq_safe API +could be called from either IRQ or process context, it disables local +IRQ then calls MEI GSC interrupt top half. + +This change fixes A380/A770 GPU boot hang issue with RT kernel. + +Fixes: 1e3dc1d8622b ("drm/i915/gsc: add gsc as a mei auxiliary device") +Tested-by: Furong Zhou +Suggested-by: Sebastian Andrzej Siewior +Acked-by: Sebastian Andrzej Siewior +Signed-off-by: Junxiao Chang +Link: https://lore.kernel.org/r/20250425151108.643649-1-junxiao.chang@intel.com +Reviewed-by: Rodrigo Vivi +Signed-off-by: Rodrigo Vivi +(cherry picked from commit dccf655f69002d496a527ba441b4f008aa5bebbf) +Signed-off-by: Joonas Lahtinen +Signed-off-by: Sasha Levin +--- + drivers/gpu/drm/i915/gt/intel_gsc.c | 2 +- + 1 file changed, 1 insertion(+), 1 deletion(-) + +diff --git a/drivers/gpu/drm/i915/gt/intel_gsc.c b/drivers/gpu/drm/i915/gt/intel_gsc.c +index 1e925c75fb080..c43febc862dc3 100644 +--- a/drivers/gpu/drm/i915/gt/intel_gsc.c ++++ b/drivers/gpu/drm/i915/gt/intel_gsc.c +@@ -284,7 +284,7 @@ static void gsc_irq_handler(struct intel_gt *gt, unsigned int intf_id) + if (gt->gsc.intf[intf_id].irq < 0) + return; + +- ret = generic_handle_irq(gt->gsc.intf[intf_id].irq); ++ ret = generic_handle_irq_safe(gt->gsc.intf[intf_id].irq); + if (ret) + gt_err_ratelimited(gt, "error handling GSC irq: %d\n", ret); + } +-- +2.39.5 + diff --git a/queue-6.15/drm-i915-gt-fix-timeline-left-held-on-vma-alloc-erro.patch b/queue-6.15/drm-i915-gt-fix-timeline-left-held-on-vma-alloc-erro.patch new file mode 100644 index 0000000000..7194e126a3 --- /dev/null +++ b/queue-6.15/drm-i915-gt-fix-timeline-left-held-on-vma-alloc-erro.patch @@ -0,0 +1,128 @@ +From ae2360644eaafc3db578f90807bcff6434ffc9b3 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Wed, 11 Jun 2025 12:42:13 +0200 +Subject: drm/i915/gt: Fix timeline left held on VMA alloc error + +From: Janusz Krzysztofik + +[ Upstream commit a5aa7bc1fca78c7fa127d9e33aa94a0c9066c1d6 ] + +The following error has been reported sporadically by CI when a test +unbinds the i915 driver on a ring submission platform: + +<4> [239.330153] ------------[ cut here ]------------ +<4> [239.330166] i915 0000:00:02.0: [drm] drm_WARN_ON(dev_priv->mm.shrink_count) +<4> [239.330196] WARNING: CPU: 1 PID: 18570 at drivers/gpu/drm/i915/i915_gem.c:1309 i915_gem_cleanup_early+0x13e/0x150 [i915] +... +<4> [239.330640] RIP: 0010:i915_gem_cleanup_early+0x13e/0x150 [i915] +... +<4> [239.330942] Call Trace: +<4> [239.330944] +<4> [239.330949] i915_driver_late_release+0x2b/0xa0 [i915] +<4> [239.331202] i915_driver_release+0x86/0xa0 [i915] +<4> [239.331482] devm_drm_dev_init_release+0x61/0x90 +<4> [239.331494] devm_action_release+0x15/0x30 +<4> [239.331504] release_nodes+0x3d/0x120 +<4> [239.331517] devres_release_all+0x96/0xd0 +<4> [239.331533] device_unbind_cleanup+0x12/0x80 +<4> [239.331543] device_release_driver_internal+0x23a/0x280 +<4> [239.331550] ? bus_find_device+0xa5/0xe0 +<4> [239.331563] device_driver_detach+0x14/0x20 +... +<4> [357.719679] ---[ end trace 0000000000000000 ]--- + +If the test also unloads the i915 module then that's followed with: + +<3> [357.787478] ============================================================================= +<3> [357.788006] BUG i915_vma (Tainted: G U W N ): Objects remaining on __kmem_cache_shutdown() +<3> [357.788031] ----------------------------------------------------------------------------- +<3> [357.788204] Object 0xffff888109e7f480 @offset=29824 +<3> [357.788670] Allocated in i915_vma_instance+0xee/0xc10 [i915] age=292729 cpu=4 pid=2244 +<4> [357.788994] i915_vma_instance+0xee/0xc10 [i915] +<4> [357.789290] init_status_page+0x7b/0x420 [i915] +<4> [357.789532] intel_engines_init+0x1d8/0x980 [i915] +<4> [357.789772] intel_gt_init+0x175/0x450 [i915] +<4> [357.790014] i915_gem_init+0x113/0x340 [i915] +<4> [357.790281] i915_driver_probe+0x847/0xed0 [i915] +<4> [357.790504] i915_pci_probe+0xe6/0x220 [i915] +... + +Closer analysis of CI results history has revealed a dependency of the +error on a few IGT tests, namely: +- igt@api_intel_allocator@fork-simple-stress-signal, +- igt@api_intel_allocator@two-level-inception-interruptible, +- igt@gem_linear_blits@interruptible, +- igt@prime_mmap_coherency@ioctl-errors, +which invisibly trigger the issue, then exhibited with first driver unbind +attempt. + +All of the above tests perform actions which are actively interrupted with +signals. Further debugging has allowed to narrow that scope down to +DRM_IOCTL_I915_GEM_EXECBUFFER2, and ring_context_alloc(), specific to ring +submission, in particular. + +If successful then that function, or its execlists or GuC submission +equivalent, is supposed to be called only once per GEM context engine, +followed by raise of a flag that prevents the function from being called +again. The function is expected to unwind its internal errors itself, so +it may be safely called once more after it returns an error. + +In case of ring submission, the function first gets a reference to the +engine's legacy timeline and then allocates a VMA. If the VMA allocation +fails, e.g. when i915_vma_instance() called from inside is interrupted +with a signal, then ring_context_alloc() fails, leaving the timeline held +referenced. On next I915_GEM_EXECBUFFER2 IOCTL, another reference to the +timeline is got, and only that last one is put on successful completion. +As a consequence, the legacy timeline, with its underlying engine status +page's VMA object, is still held and not released on driver unbind. + +Get the legacy timeline only after successful allocation of the context +engine's VMA. + +v2: Add a note on other submission methods (Krzysztof Karas): + Both execlists and GuC submission use lrc_alloc() which seems free + from a similar issue. + +Fixes: 75d0a7f31eec ("drm/i915: Lift timeline into intel_context") +Closes: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/12061 +Cc: Chris Wilson +Cc: Matthew Auld +Cc: Krzysztof Karas +Reviewed-by: Sebastian Brzezinka +Reviewed-by: Krzysztof Niemiec +Signed-off-by: Janusz Krzysztofik +Reviewed-by: Nitin Gote +Reviewed-by: Andi Shyti +Signed-off-by: Andi Shyti +Link: https://lore.kernel.org/r/20250611104352.1014011-2-janusz.krzysztofik@linux.intel.com +(cherry picked from commit cc43422b3cc79eacff4c5a8ba0d224688ca9dd4f) +Signed-off-by: Joonas Lahtinen +Signed-off-by: Sasha Levin +--- + drivers/gpu/drm/i915/gt/intel_ring_submission.c | 3 ++- + 1 file changed, 2 insertions(+), 1 deletion(-) + +diff --git a/drivers/gpu/drm/i915/gt/intel_ring_submission.c b/drivers/gpu/drm/i915/gt/intel_ring_submission.c +index 6e9977b2d1802..44a4b6723582f 100644 +--- a/drivers/gpu/drm/i915/gt/intel_ring_submission.c ++++ b/drivers/gpu/drm/i915/gt/intel_ring_submission.c +@@ -604,7 +604,6 @@ static int ring_context_alloc(struct intel_context *ce) + /* One ringbuffer to rule them all */ + GEM_BUG_ON(!engine->legacy.ring); + ce->ring = engine->legacy.ring; +- ce->timeline = intel_timeline_get(engine->legacy.timeline); + + GEM_BUG_ON(ce->state); + if (engine->context_size) { +@@ -617,6 +616,8 @@ static int ring_context_alloc(struct intel_context *ce) + ce->state = vma; + } + ++ ce->timeline = intel_timeline_get(engine->legacy.timeline); ++ + return 0; + } + +-- +2.39.5 + diff --git a/queue-6.15/drm-i915-selftests-change-mock_request-to-return-err.patch b/queue-6.15/drm-i915-selftests-change-mock_request-to-return-err.patch new file mode 100644 index 0000000000..21178f3eee --- /dev/null +++ b/queue-6.15/drm-i915-selftests-change-mock_request-to-return-err.patch @@ -0,0 +1,107 @@ +From 1d4ef85f62f6cf79443677c8594a7109ae6ca0ad Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Wed, 25 Jun 2025 10:21:58 -0500 +Subject: drm/i915/selftests: Change mock_request() to return error pointers + +From: Dan Carpenter + +[ Upstream commit caa7c7a76b78ce41d347003f84975125383e6b59 ] + +There was an error pointer vs NULL bug in __igt_breadcrumbs_smoketest(). +The __mock_request_alloc() function implements the +smoketest->request_alloc() function pointer. It was supposed to return +error pointers, but it propogates the NULL return from mock_request() +so in the event of a failure, it would lead to a NULL pointer +dereference. + +To fix this, change the mock_request() function to return error pointers +and update all the callers to expect that. + +Fixes: 52c0fdb25c7c ("drm/i915: Replace global breadcrumbs with per-context interrupt tracking") +Signed-off-by: Dan Carpenter +Reviewed-by: Rodrigo Vivi +Link: https://lore.kernel.org/r/685c1417.050a0220.696f5.5c05@mx.google.com +Signed-off-by: Rodrigo Vivi +(cherry picked from commit 778fa8ad5f0f23397d045c7ebca048ce8def1c43) +Signed-off-by: Joonas Lahtinen +Signed-off-by: Sasha Levin +--- + drivers/gpu/drm/i915/selftests/i915_request.c | 20 +++++++++---------- + drivers/gpu/drm/i915/selftests/mock_request.c | 2 +- + 2 files changed, 11 insertions(+), 11 deletions(-) + +diff --git a/drivers/gpu/drm/i915/selftests/i915_request.c b/drivers/gpu/drm/i915/selftests/i915_request.c +index 88870844b5bd9..2fb7a9e7efec6 100644 +--- a/drivers/gpu/drm/i915/selftests/i915_request.c ++++ b/drivers/gpu/drm/i915/selftests/i915_request.c +@@ -73,8 +73,8 @@ static int igt_add_request(void *arg) + /* Basic preliminary test to create a request and let it loose! */ + + request = mock_request(rcs0(i915)->kernel_context, HZ / 10); +- if (!request) +- return -ENOMEM; ++ if (IS_ERR(request)) ++ return PTR_ERR(request); + + i915_request_add(request); + +@@ -91,8 +91,8 @@ static int igt_wait_request(void *arg) + /* Submit a request, then wait upon it */ + + request = mock_request(rcs0(i915)->kernel_context, T); +- if (!request) +- return -ENOMEM; ++ if (IS_ERR(request)) ++ return PTR_ERR(request); + + i915_request_get(request); + +@@ -160,8 +160,8 @@ static int igt_fence_wait(void *arg) + /* Submit a request, treat it as a fence and wait upon it */ + + request = mock_request(rcs0(i915)->kernel_context, T); +- if (!request) +- return -ENOMEM; ++ if (IS_ERR(request)) ++ return PTR_ERR(request); + + if (dma_fence_wait_timeout(&request->fence, false, T) != -ETIME) { + pr_err("fence wait success before submit (expected timeout)!\n"); +@@ -219,8 +219,8 @@ static int igt_request_rewind(void *arg) + GEM_BUG_ON(IS_ERR(ce)); + request = mock_request(ce, 2 * HZ); + intel_context_put(ce); +- if (!request) { +- err = -ENOMEM; ++ if (IS_ERR(request)) { ++ err = PTR_ERR(request); + goto err_context_0; + } + +@@ -237,8 +237,8 @@ static int igt_request_rewind(void *arg) + GEM_BUG_ON(IS_ERR(ce)); + vip = mock_request(ce, 0); + intel_context_put(ce); +- if (!vip) { +- err = -ENOMEM; ++ if (IS_ERR(vip)) { ++ err = PTR_ERR(vip); + goto err_context_1; + } + +diff --git a/drivers/gpu/drm/i915/selftests/mock_request.c b/drivers/gpu/drm/i915/selftests/mock_request.c +index 09f747228dff5..1b0cf073e9643 100644 +--- a/drivers/gpu/drm/i915/selftests/mock_request.c ++++ b/drivers/gpu/drm/i915/selftests/mock_request.c +@@ -35,7 +35,7 @@ mock_request(struct intel_context *ce, unsigned long delay) + /* NB the i915->requests slab cache is enlarged to fit mock_request */ + request = intel_context_create_request(ce); + if (IS_ERR(request)) +- return NULL; ++ return request; + + request->mock.delay = delay; + return request; +-- +2.39.5 + diff --git a/queue-6.15/drm-msm-fix-a-fence-leak-in-submit-error-path.patch b/queue-6.15/drm-msm-fix-a-fence-leak-in-submit-error-path.patch new file mode 100644 index 0000000000..c2c58064e8 --- /dev/null +++ b/queue-6.15/drm-msm-fix-a-fence-leak-in-submit-error-path.patch @@ -0,0 +1,45 @@ +From f5bba720698fbc21c6c8d6f72cd368187ed2e213 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Wed, 14 May 2025 09:33:32 -0700 +Subject: drm/msm: Fix a fence leak in submit error path + +From: Rob Clark + +[ Upstream commit 5d319f75ccf7f0927425a7545aa1a22b3eedc189 ] + +In error paths, we could unref the submit without calling +drm_sched_entity_push_job(), so msm_job_free() will never get +called. Since drm_sched_job_cleanup() will NULL out the +s_fence, we can use that to detect this case. + +Signed-off-by: Rob Clark +Patchwork: https://patchwork.freedesktop.org/patch/653584/ +Signed-off-by: Rob Clark +Signed-off-by: Sasha Levin +--- + drivers/gpu/drm/msm/msm_gem_submit.c | 9 +++++++++ + 1 file changed, 9 insertions(+) + +diff --git a/drivers/gpu/drm/msm/msm_gem_submit.c b/drivers/gpu/drm/msm/msm_gem_submit.c +index 3e9aa2cc38ef9..b2aeaecaa39b3 100644 +--- a/drivers/gpu/drm/msm/msm_gem_submit.c ++++ b/drivers/gpu/drm/msm/msm_gem_submit.c +@@ -85,6 +85,15 @@ void __msm_gem_submit_destroy(struct kref *kref) + container_of(kref, struct msm_gem_submit, ref); + unsigned i; + ++ /* ++ * In error paths, we could unref the submit without calling ++ * drm_sched_entity_push_job(), so msm_job_free() will never ++ * get called. Since drm_sched_job_cleanup() will NULL out ++ * s_fence, we can use that to detect this case. ++ */ ++ if (submit->base.s_fence) ++ drm_sched_job_cleanup(&submit->base); ++ + if (submit->fence_id) { + spin_lock(&submit->queue->idr_lock); + idr_remove(&submit->queue->fence_idr, submit->fence_id); +-- +2.39.5 + diff --git a/queue-6.15/drm-msm-fix-another-leak-in-the-submit-error-path.patch b/queue-6.15/drm-msm-fix-another-leak-in-the-submit-error-path.patch new file mode 100644 index 0000000000..a3c6b9a6cf --- /dev/null +++ b/queue-6.15/drm-msm-fix-another-leak-in-the-submit-error-path.patch @@ -0,0 +1,57 @@ +From 8f08787b49c743fe58cf596fa26923dc03a5f2ce Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Wed, 14 May 2025 09:33:33 -0700 +Subject: drm/msm: Fix another leak in the submit error path + +From: Rob Clark + +[ Upstream commit f681c2aa8676a890eacc84044717ab0fd26e058f ] + +put_unused_fd() doesn't free the installed file, if we've already done +fd_install(). So we need to also free the sync_file. + +Signed-off-by: Rob Clark +Patchwork: https://patchwork.freedesktop.org/patch/653583/ +Signed-off-by: Rob Clark +Signed-off-by: Sasha Levin +--- + drivers/gpu/drm/msm/msm_gem_submit.c | 8 ++++++-- + 1 file changed, 6 insertions(+), 2 deletions(-) + +diff --git a/drivers/gpu/drm/msm/msm_gem_submit.c b/drivers/gpu/drm/msm/msm_gem_submit.c +index b2aeaecaa39b3..d4f71bb54e84c 100644 +--- a/drivers/gpu/drm/msm/msm_gem_submit.c ++++ b/drivers/gpu/drm/msm/msm_gem_submit.c +@@ -658,6 +658,7 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data, + struct msm_ringbuffer *ring; + struct msm_submit_post_dep *post_deps = NULL; + struct drm_syncobj **syncobjs_to_reset = NULL; ++ struct sync_file *sync_file = NULL; + int out_fence_fd = -1; + unsigned i; + int ret; +@@ -867,7 +868,7 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data, + } + + if (ret == 0 && args->flags & MSM_SUBMIT_FENCE_FD_OUT) { +- struct sync_file *sync_file = sync_file_create(submit->user_fence); ++ sync_file = sync_file_create(submit->user_fence); + if (!sync_file) { + ret = -ENOMEM; + } else { +@@ -901,8 +902,11 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data, + out_unlock: + mutex_unlock(&queue->lock); + out_post_unlock: +- if (ret && (out_fence_fd >= 0)) ++ if (ret && (out_fence_fd >= 0)) { + put_unused_fd(out_fence_fd); ++ if (sync_file) ++ fput(sync_file->file); ++ } + + if (!IS_ERR_OR_NULL(submit)) { + msm_gem_submit_put(submit); +-- +2.39.5 + diff --git a/queue-6.15/drm-vmwgfx-fix-guests-running-with-tdx-sev.patch b/queue-6.15/drm-vmwgfx-fix-guests-running-with-tdx-sev.patch new file mode 100644 index 0000000000..1becc2610f --- /dev/null +++ b/queue-6.15/drm-vmwgfx-fix-guests-running-with-tdx-sev.patch @@ -0,0 +1,53 @@ +From 2e7f386f232cea28192f42a11bab658cb24bb046 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Wed, 18 Jun 2025 15:29:26 -0400 +Subject: drm/vmwgfx: Fix guests running with TDX/SEV + +From: Marko Kiiskila + +[ Upstream commit 7dfede7d7edd18c0c91ca854cde8eaaf4ccf97ea ] + +Commit 81256a50aa0f ("x86/mm: Make memremap(MEMREMAP_WB) map memory as +encrypted by default") changed the default behavior of +memremap(MEMREMAP_WB) and started mapping memory as encrypted. +The driver requires the fifo memory to be decrypted to communicate with +the host but was relaying on the old default behavior of +memremap(MEMREMAP_WB) and thus broke. + +Fix it by explicitly specifying the desired behavior and passing +MEMREMAP_DEC to memremap. + +Fixes: 81256a50aa0f ("x86/mm: Make memremap(MEMREMAP_WB) map memory as encrypted by default") +Signed-off-by: Marko Kiiskila +Signed-off-by: Zack Rusin +Cc: Kirill A. Shutemov +Cc: Ingo Molnar +Cc: Andrew Morton +Cc: Dave Hansen +Cc: Linus Torvalds +Cc: Peter Zijlstra +Cc: linux-mm@kvack.org +Cc: linux-kernel@vger.kernel.org +Acked-by: Kirill A. Shutemov +Link: https://lore.kernel.org/r/20250618192926.1092450-1-zack.rusin@broadcom.com +Signed-off-by: Sasha Levin +--- + drivers/gpu/drm/vmwgfx/vmwgfx_drv.c | 2 +- + 1 file changed, 1 insertion(+), 1 deletion(-) + +diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c +index 0f32471c85332..55c822a61b9ad 100644 +--- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c ++++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c +@@ -769,7 +769,7 @@ static int vmw_setup_pci_resources(struct vmw_private *dev, + dev->fifo_mem = devm_memremap(dev->drm.dev, + fifo_start, + fifo_size, +- MEMREMAP_WB); ++ MEMREMAP_WB | MEMREMAP_DEC); + + if (IS_ERR(dev->fifo_mem)) { + drm_err(&dev->drm, +-- +2.39.5 + diff --git a/queue-6.15/drm-xe-allow-dropping-kunit-dependency-as-built-in.patch b/queue-6.15/drm-xe-allow-dropping-kunit-dependency-as-built-in.patch new file mode 100644 index 0000000000..356bdfdbbc --- /dev/null +++ b/queue-6.15/drm-xe-allow-dropping-kunit-dependency-as-built-in.patch @@ -0,0 +1,51 @@ +From 6ec28634146caff107a000bdab8cee5c15add3d5 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Fri, 27 Jun 2025 13:30:35 -0700 +Subject: drm/xe: Allow dropping kunit dependency as built-in +MIME-Version: 1.0 +Content-Type: text/plain; charset=UTF-8 +Content-Transfer-Encoding: 8bit + +From: Harry Austen + +[ Upstream commit aa18d5769fcafe645a3ba01a9a69dde4f8dc8cc3 ] + +Fix Kconfig symbol dependency on KUNIT, which isn't actually required +for XE to be built-in. However, if KUNIT is enabled, it must be built-in +too. + +Fixes: 08987a8b6820 ("drm/xe: Fix build with KUNIT=m") +Cc: Lucas De Marchi +Cc: Thomas Hellström +Cc: Jani Nikula +Cc: Maarten Lankhorst +Signed-off-by: Harry Austen +Reviewed-by: Lucas De Marchi +Acked-by: Randy Dunlap +Tested-by: Randy Dunlap +Link: https://lore.kernel.org/r/20250627-xe-kunit-v2-2-756fe5cd56cf@intel.com +Signed-off-by: Lucas De Marchi +(cherry picked from commit a559434880b320b83733d739733250815aecf1b0) +Signed-off-by: Lucas De Marchi +Signed-off-by: Sasha Levin +--- + drivers/gpu/drm/xe/Kconfig | 3 ++- + 1 file changed, 2 insertions(+), 1 deletion(-) + +diff --git a/drivers/gpu/drm/xe/Kconfig b/drivers/gpu/drm/xe/Kconfig +index 9a46aafcb33ba..0dc3512876b3a 100644 +--- a/drivers/gpu/drm/xe/Kconfig ++++ b/drivers/gpu/drm/xe/Kconfig +@@ -1,7 +1,8 @@ + # SPDX-License-Identifier: GPL-2.0-only + config DRM_XE + tristate "Intel Xe Graphics" +- depends on DRM && PCI && MMU && (m || (y && KUNIT=y)) ++ depends on DRM && PCI && MMU ++ depends on KUNIT || !KUNIT + depends on INTEL_VSEC || !INTEL_VSEC + depends on X86_PLATFORM_DEVICES || !(X86 && ACPI) + select INTERVAL_TREE +-- +2.39.5 + diff --git a/queue-6.15/drm-xe-bmg-update-wa_14022085890.patch b/queue-6.15/drm-xe-bmg-update-wa_14022085890.patch new file mode 100644 index 0000000000..dfe0ed32f2 --- /dev/null +++ b/queue-6.15/drm-xe-bmg-update-wa_14022085890.patch @@ -0,0 +1,75 @@ +From 821299eeae5d5f476f4663f1be12b3782ff8b86e Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Thu, 12 Jun 2025 00:09:02 -0700 +Subject: drm/xe/bmg: Update Wa_14022085890 + +From: Vinay Belgaumkar + +[ Upstream commit a5c7dcdd969f2248cc91d65e5ac852859fc8dac2 ] + +Set GT min frequency to 1200Mhz once driver load is complete. + +v2: Review comments (Rodrigo) +v3: Apply Wa earlier so user_req_min is not clobbered. +v4: Apply to all GTs (Lucas) + +Cc: Matt Roper +Cc: Rodrigo Vivi +Signed-off-by: Vinay Belgaumkar +Reviewed-by: Stuart Summers +Link: https://lore.kernel.org/r/20250612-wa-14022085890-v4-3-94ba5dcc1e30@intel.com +Signed-off-by: Lucas De Marchi +(cherry picked from commit bdde16c9ac5cb56ad2ee19792222fa1853577af7) +Signed-off-by: Lucas De Marchi +Stable-dep-of: 84c0b4a00610 ("drm/xe/bmg: Update Wa_22019338487") +Signed-off-by: Sasha Levin +--- + drivers/gpu/drm/xe/xe_guc_pc.c | 5 +++++ + drivers/gpu/drm/xe/xe_wa_oob.rules | 4 ++++ + 2 files changed, 9 insertions(+) + +diff --git a/drivers/gpu/drm/xe/xe_guc_pc.c b/drivers/gpu/drm/xe/xe_guc_pc.c +index 23a4c525c03bf..28b97a2c14e3b 100644 +--- a/drivers/gpu/drm/xe/xe_guc_pc.c ++++ b/drivers/gpu/drm/xe/xe_guc_pc.c +@@ -52,6 +52,7 @@ + + #define LNL_MERT_FREQ_CAP 800 + #define BMG_MERT_FREQ_CAP 2133 ++#define BMG_MIN_FREQ 1200 + + #define SLPC_RESET_TIMEOUT_MS 5 /* roughly 5ms, but no need for precision */ + #define SLPC_RESET_EXTENDED_TIMEOUT_MS 1000 /* To be used only at pc_start */ +@@ -817,6 +818,7 @@ void xe_guc_pc_init_early(struct xe_guc_pc *pc) + + static int pc_adjust_freq_bounds(struct xe_guc_pc *pc) + { ++ struct xe_tile *tile = gt_to_tile(pc_to_gt(pc)); + int ret; + + lockdep_assert_held(&pc->freq_lock); +@@ -843,6 +845,9 @@ static int pc_adjust_freq_bounds(struct xe_guc_pc *pc) + if (pc_get_min_freq(pc) > pc->rp0_freq) + ret = pc_set_min_freq(pc, pc->rp0_freq); + ++ if (XE_WA(tile->primary_gt, 14022085890)) ++ ret = pc_set_min_freq(pc, max(BMG_MIN_FREQ, pc_get_min_freq(pc))); ++ + out: + return ret; + } +diff --git a/drivers/gpu/drm/xe/xe_wa_oob.rules b/drivers/gpu/drm/xe/xe_wa_oob.rules +index 9efc5accd43d1..320766f6c5dff 100644 +--- a/drivers/gpu/drm/xe/xe_wa_oob.rules ++++ b/drivers/gpu/drm/xe/xe_wa_oob.rules +@@ -59,3 +59,7 @@ no_media_l3 MEDIA_VERSION(3000) + MEDIA_VERSION_RANGE(1301, 3000) + 16026508708 GRAPHICS_VERSION_RANGE(1200, 3001) + MEDIA_VERSION_RANGE(1300, 3000) ++ ++# SoC workaround - currently applies to all platforms with the following ++# primary GT GMDID ++14022085890 GRAPHICS_VERSION(2001) +-- +2.39.5 + diff --git a/queue-6.15/drm-xe-bmg-update-wa_22019338487.patch b/queue-6.15/drm-xe-bmg-update-wa_22019338487.patch new file mode 100644 index 0000000000..ab876917fd --- /dev/null +++ b/queue-6.15/drm-xe-bmg-update-wa_22019338487.patch @@ -0,0 +1,266 @@ +From 0d9d9a01a8656e6c5c319eec853df1ef8b65b805 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Wed, 18 Jun 2025 11:50:01 -0700 +Subject: drm/xe/bmg: Update Wa_22019338487 + +From: Vinay Belgaumkar + +[ Upstream commit 84c0b4a00610afbde650fdb8ad6db0424f7b2cc3 ] + +Limit GT max frequency to 2600MHz and wait for frequency to reduce +before proceeding with a transient flush. This is really only needed for +the transient flush: if L2 flush is needed due to 16023588340 then +there's no need to do this additional wait since we are already using +the bigger hammer. + +v2: Use generic names, ensure user set max frequency requests wait +for flush to complete (Rodrigo) +v3: + - User requests wait via wait_var_event_timeout (Lucas) + - Close races on flush + user requests (Lucas) + - Fix xe_guc_pc_remove_flush_freq_limit() being called on last gt + rather than root gt (Lucas) +v4: + - Only apply the freq reducing part if a TDF is needed: L2 flush trumps + the need for waiting a lower frequency + +Fixes: aaa08078e725 ("drm/xe/bmg: Apply Wa_22019338487") +Reviewed-by: Rodrigo Vivi +Signed-off-by: Vinay Belgaumkar +Link: https://lore.kernel.org/r/20250618-wa-22019338487-v5-4-b888388477f2@intel.com +Signed-off-by: Lucas De Marchi +(cherry picked from commit deea6a7d6d803d6bb874a3e6f1b312e560e6c6df) +Signed-off-by: Lucas De Marchi +Signed-off-by: Sasha Levin +--- + drivers/gpu/drm/xe/xe_device.c | 8 +- + drivers/gpu/drm/xe/xe_guc_pc.c | 125 +++++++++++++++++++++++++++ + drivers/gpu/drm/xe/xe_guc_pc.h | 2 + + drivers/gpu/drm/xe/xe_guc_pc_types.h | 2 + + 4 files changed, 135 insertions(+), 2 deletions(-) + +diff --git a/drivers/gpu/drm/xe/xe_device.c b/drivers/gpu/drm/xe/xe_device.c +index 38fdddd7262aa..f3123914b1abf 100644 +--- a/drivers/gpu/drm/xe/xe_device.c ++++ b/drivers/gpu/drm/xe/xe_device.c +@@ -38,6 +38,7 @@ + #include "xe_gt_printk.h" + #include "xe_gt_sriov_vf.h" + #include "xe_guc.h" ++#include "xe_guc_pc.h" + #include "xe_hw_engine_group.h" + #include "xe_hwmon.h" + #include "xe_irq.h" +@@ -1057,11 +1058,14 @@ void xe_device_td_flush(struct xe_device *xe) + return; + + root_gt = xe_root_mmio_gt(xe); +- if (XE_WA(root_gt, 16023588340)) ++ if (XE_WA(root_gt, 16023588340)) { + /* A transient flush is not sufficient: flush the L2 */ + xe_device_l2_flush(xe); +- else ++ } else { ++ xe_guc_pc_apply_flush_freq_limit(&root_gt->uc.guc.pc); + tdf_request_sync(xe); ++ xe_guc_pc_remove_flush_freq_limit(&root_gt->uc.guc.pc); ++ } + } + + u32 xe_device_ccs_bytes(struct xe_device *xe, u64 size) +diff --git a/drivers/gpu/drm/xe/xe_guc_pc.c b/drivers/gpu/drm/xe/xe_guc_pc.c +index 28b97a2c14e3b..1c7b044413f26 100644 +--- a/drivers/gpu/drm/xe/xe_guc_pc.c ++++ b/drivers/gpu/drm/xe/xe_guc_pc.c +@@ -7,7 +7,9 @@ + + #include + #include ++#include + #include ++#include + + #include + #include +@@ -53,9 +55,11 @@ + #define LNL_MERT_FREQ_CAP 800 + #define BMG_MERT_FREQ_CAP 2133 + #define BMG_MIN_FREQ 1200 ++#define BMG_MERT_FLUSH_FREQ_CAP 2600 + + #define SLPC_RESET_TIMEOUT_MS 5 /* roughly 5ms, but no need for precision */ + #define SLPC_RESET_EXTENDED_TIMEOUT_MS 1000 /* To be used only at pc_start */ ++#define SLPC_ACT_FREQ_TIMEOUT_MS 100 + + /** + * DOC: GuC Power Conservation (PC) +@@ -143,6 +147,36 @@ static int wait_for_pc_state(struct xe_guc_pc *pc, + return -ETIMEDOUT; + } + ++static int wait_for_flush_complete(struct xe_guc_pc *pc) ++{ ++ const unsigned long timeout = msecs_to_jiffies(30); ++ ++ if (!wait_var_event_timeout(&pc->flush_freq_limit, ++ !atomic_read(&pc->flush_freq_limit), ++ timeout)) ++ return -ETIMEDOUT; ++ ++ return 0; ++} ++ ++static int wait_for_act_freq_limit(struct xe_guc_pc *pc, u32 freq) ++{ ++ int timeout_us = SLPC_ACT_FREQ_TIMEOUT_MS * USEC_PER_MSEC; ++ int slept, wait = 10; ++ ++ for (slept = 0; slept < timeout_us;) { ++ if (xe_guc_pc_get_act_freq(pc) <= freq) ++ return 0; ++ ++ usleep_range(wait, wait << 1); ++ slept += wait; ++ wait <<= 1; ++ if (slept + wait > timeout_us) ++ wait = timeout_us - slept; ++ } ++ ++ return -ETIMEDOUT; ++} + static int pc_action_reset(struct xe_guc_pc *pc) + { + struct xe_guc_ct *ct = pc_to_ct(pc); +@@ -673,6 +707,11 @@ static int xe_guc_pc_set_max_freq_locked(struct xe_guc_pc *pc, u32 freq) + */ + int xe_guc_pc_set_max_freq(struct xe_guc_pc *pc, u32 freq) + { ++ if (XE_WA(pc_to_gt(pc), 22019338487)) { ++ if (wait_for_flush_complete(pc) != 0) ++ return -EAGAIN; ++ } ++ + guard(mutex)(&pc->freq_lock); + + return xe_guc_pc_set_max_freq_locked(pc, freq); +@@ -873,6 +912,92 @@ static int pc_adjust_requested_freq(struct xe_guc_pc *pc) + return ret; + } + ++static bool needs_flush_freq_limit(struct xe_guc_pc *pc) ++{ ++ struct xe_gt *gt = pc_to_gt(pc); ++ ++ return XE_WA(gt, 22019338487) && ++ pc->rp0_freq > BMG_MERT_FLUSH_FREQ_CAP; ++} ++ ++/** ++ * xe_guc_pc_apply_flush_freq_limit() - Limit max GT freq during L2 flush ++ * @pc: the xe_guc_pc object ++ * ++ * As per the WA, reduce max GT frequency during L2 cache flush ++ */ ++void xe_guc_pc_apply_flush_freq_limit(struct xe_guc_pc *pc) ++{ ++ struct xe_gt *gt = pc_to_gt(pc); ++ u32 max_freq; ++ int ret; ++ ++ if (!needs_flush_freq_limit(pc)) ++ return; ++ ++ guard(mutex)(&pc->freq_lock); ++ ++ ret = xe_guc_pc_get_max_freq_locked(pc, &max_freq); ++ if (!ret && max_freq > BMG_MERT_FLUSH_FREQ_CAP) { ++ ret = pc_set_max_freq(pc, BMG_MERT_FLUSH_FREQ_CAP); ++ if (ret) { ++ xe_gt_err_once(gt, "Failed to cap max freq on flush to %u, %pe\n", ++ BMG_MERT_FLUSH_FREQ_CAP, ERR_PTR(ret)); ++ return; ++ } ++ ++ atomic_set(&pc->flush_freq_limit, 1); ++ ++ /* ++ * If user has previously changed max freq, stash that value to ++ * restore later, otherwise use the current max. New user ++ * requests wait on flush. ++ */ ++ if (pc->user_requested_max != 0) ++ pc->stashed_max_freq = pc->user_requested_max; ++ else ++ pc->stashed_max_freq = max_freq; ++ } ++ ++ /* ++ * Wait for actual freq to go below the flush cap: even if the previous ++ * max was below cap, the current one might still be above it ++ */ ++ ret = wait_for_act_freq_limit(pc, BMG_MERT_FLUSH_FREQ_CAP); ++ if (ret) ++ xe_gt_err_once(gt, "Actual freq did not reduce to %u, %pe\n", ++ BMG_MERT_FLUSH_FREQ_CAP, ERR_PTR(ret)); ++} ++ ++/** ++ * xe_guc_pc_remove_flush_freq_limit() - Remove max GT freq limit after L2 flush completes. ++ * @pc: the xe_guc_pc object ++ * ++ * Retrieve the previous GT max frequency value. ++ */ ++void xe_guc_pc_remove_flush_freq_limit(struct xe_guc_pc *pc) ++{ ++ struct xe_gt *gt = pc_to_gt(pc); ++ int ret = 0; ++ ++ if (!needs_flush_freq_limit(pc)) ++ return; ++ ++ if (!atomic_read(&pc->flush_freq_limit)) ++ return; ++ ++ mutex_lock(&pc->freq_lock); ++ ++ ret = pc_set_max_freq(>->uc.guc.pc, pc->stashed_max_freq); ++ if (ret) ++ xe_gt_err_once(gt, "Failed to restore max freq %u:%d", ++ pc->stashed_max_freq, ret); ++ ++ atomic_set(&pc->flush_freq_limit, 0); ++ mutex_unlock(&pc->freq_lock); ++ wake_up_var(&pc->flush_freq_limit); ++} ++ + static int pc_set_mert_freq_cap(struct xe_guc_pc *pc) + { + int ret = 0; +diff --git a/drivers/gpu/drm/xe/xe_guc_pc.h b/drivers/gpu/drm/xe/xe_guc_pc.h +index 39102b79602fd..0302c7426ccde 100644 +--- a/drivers/gpu/drm/xe/xe_guc_pc.h ++++ b/drivers/gpu/drm/xe/xe_guc_pc.h +@@ -37,5 +37,7 @@ u64 xe_guc_pc_mc6_residency(struct xe_guc_pc *pc); + void xe_guc_pc_init_early(struct xe_guc_pc *pc); + int xe_guc_pc_restore_stashed_freq(struct xe_guc_pc *pc); + void xe_guc_pc_raise_unslice(struct xe_guc_pc *pc); ++void xe_guc_pc_apply_flush_freq_limit(struct xe_guc_pc *pc); ++void xe_guc_pc_remove_flush_freq_limit(struct xe_guc_pc *pc); + + #endif /* _XE_GUC_PC_H_ */ +diff --git a/drivers/gpu/drm/xe/xe_guc_pc_types.h b/drivers/gpu/drm/xe/xe_guc_pc_types.h +index 2978ac9a249b5..c02053948a579 100644 +--- a/drivers/gpu/drm/xe/xe_guc_pc_types.h ++++ b/drivers/gpu/drm/xe/xe_guc_pc_types.h +@@ -15,6 +15,8 @@ + struct xe_guc_pc { + /** @bo: GGTT buffer object that is shared with GuC PC */ + struct xe_bo *bo; ++ /** @flush_freq_limit: 1 when max freq changes are limited by driver */ ++ atomic_t flush_freq_limit; + /** @rp0_freq: HW RP0 frequency - The Maximum one */ + u32 rp0_freq; + /** @rpa_freq: HW RPa frequency - The Achievable one */ +-- +2.39.5 + diff --git a/queue-6.15/drm-xe-fix-out-of-bounds-field-write-in-mi_store_dat.patch b/queue-6.15/drm-xe-fix-out-of-bounds-field-write-in-mi_store_dat.patch new file mode 100644 index 0000000000..dc0e9c2274 --- /dev/null +++ b/queue-6.15/drm-xe-fix-out-of-bounds-field-write-in-mi_store_dat.patch @@ -0,0 +1,92 @@ +From 1c25565acf3d654563e843becea17df8551406ef Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Thu, 12 Jun 2025 22:46:20 +0000 +Subject: drm/xe: Fix out-of-bounds field write in MI_STORE_DATA_IMM + +From: Jia Yao + +[ Upstream commit 2d5cff2b4bc567dcaad7ab5b46c973ba534cc062 ] + +According to Bspec, bits 0~9 of MI_STORE_DATA_IMM must not exceed 0x3FE. +The macro MI_SDI_NUM_QW(x) evaluates to 2 * x + 1, which means the +condition 2 * x + 1 <= 0x3FE must be satisfied. Therefore, the maximum +valid value for x is 0x1FE, not 0x1FF. + +v2 + - Replace 0x1fe with macro MAX_PTE_PER_SDI (Auld, Matthew & Patelczyk, Maciej) + +v3 + - Change macro MAX_PTE_PER_SDI from 0x1fe to 0x1feU (De Marchi, Lucas) + +Bspec: 60246 + +Fixes: 9c44fd5f6e8a ("drm/xe: Add migrate layer functions for SVM support") +Cc: Matthew Brost +Cc: Brian3 Nguyen +Cc: Alex Zuo +Cc: Matthew Auld +Cc: Maciej Patelczyk +Cc: Lucas De Marchi +Suggested-by: Shuicheng Lin +Signed-off-by: Jia Yao +Reviewed-by: Matthew Brost +Reviewed-by: Lucas De Marchi +Reviewed-by: Maciej Patelczyk +Link: https://lore.kernel.org/r/20250612224620.161105-1-jia.yao@intel.com +Signed-off-by: Lucas De Marchi +(cherry picked from commit c038bdba98c9f6a36378044a9d4385531a194d3e) +Signed-off-by: Lucas De Marchi +Signed-off-by: Sasha Levin +--- + drivers/gpu/drm/xe/xe_migrate.c | 18 ++++++++++-------- + 1 file changed, 10 insertions(+), 8 deletions(-) + +diff --git a/drivers/gpu/drm/xe/xe_migrate.c b/drivers/gpu/drm/xe/xe_migrate.c +index 5a3e89022c381..752f57fd515e1 100644 +--- a/drivers/gpu/drm/xe/xe_migrate.c ++++ b/drivers/gpu/drm/xe/xe_migrate.c +@@ -82,7 +82,7 @@ struct xe_migrate { + * of the instruction. Subtracting the instruction header (1 dword) and + * address (2 dwords), that leaves 0x3FD dwords (0x1FE qwords) for PTE values. + */ +-#define MAX_PTE_PER_SDI 0x1FE ++#define MAX_PTE_PER_SDI 0x1FEU + + /** + * xe_tile_migrate_exec_queue() - Get this tile's migrate exec queue. +@@ -1550,15 +1550,17 @@ static u32 pte_update_cmd_size(u64 size) + u64 entries = DIV_U64_ROUND_UP(size, XE_PAGE_SIZE); + + XE_WARN_ON(size > MAX_PREEMPTDISABLE_TRANSFER); ++ + /* + * MI_STORE_DATA_IMM command is used to update page table. Each +- * instruction can update maximumly 0x1ff pte entries. To update +- * n (n <= 0x1ff) pte entries, we need: +- * 1 dword for the MI_STORE_DATA_IMM command header (opcode etc) +- * 2 dword for the page table's physical location +- * 2*n dword for value of pte to fill (each pte entry is 2 dwords) ++ * instruction can update maximumly MAX_PTE_PER_SDI pte entries. To ++ * update n (n <= MAX_PTE_PER_SDI) pte entries, we need: ++ * ++ * - 1 dword for the MI_STORE_DATA_IMM command header (opcode etc) ++ * - 2 dword for the page table's physical location ++ * - 2*n dword for value of pte to fill (each pte entry is 2 dwords) + */ +- num_dword = (1 + 2) * DIV_U64_ROUND_UP(entries, 0x1ff); ++ num_dword = (1 + 2) * DIV_U64_ROUND_UP(entries, MAX_PTE_PER_SDI); + num_dword += entries * 2; + + return num_dword; +@@ -1574,7 +1576,7 @@ static void build_pt_update_batch_sram(struct xe_migrate *m, + + ptes = DIV_ROUND_UP(size, XE_PAGE_SIZE); + while (ptes) { +- u32 chunk = min(0x1ffU, ptes); ++ u32 chunk = min(MAX_PTE_PER_SDI, ptes); + + bb->cs[bb->len++] = MI_STORE_DATA_IMM | MI_SDI_NUM_QW(chunk); + bb->cs[bb->len++] = pt_offset; +-- +2.39.5 + diff --git a/queue-6.15/drm-xe-guc-enable-w-a-16026508708.patch b/queue-6.15/drm-xe-guc-enable-w-a-16026508708.patch new file mode 100644 index 0000000000..70715c10cf --- /dev/null +++ b/queue-6.15/drm-xe-guc-enable-w-a-16026508708.patch @@ -0,0 +1,64 @@ +From ece7fb6ecebd5edbb49bde6c52091faa1272ed1a Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Thu, 3 Apr 2025 11:56:11 -0700 +Subject: drm/xe/guc: Enable w/a 16026508708 + +From: John Harrison + +[ Upstream commit d3e8349edf7ed9eaf076ab3d9973331ccc20e26c ] + +The workaround is only relevant to SRIOV but does affect all platforms. + +Signed-off-by: John Harrison +Reviewed-by: Daniele Ceraolo Spurio +Link: https://lore.kernel.org/r/20250403185619.1555853-2-John.C.Harrison@Intel.com +Signed-off-by: Lucas De Marchi +Stable-dep-of: 84c0b4a00610 ("drm/xe/bmg: Update Wa_22019338487") +Signed-off-by: Sasha Levin +--- + drivers/gpu/drm/xe/abi/guc_klvs_abi.h | 1 + + drivers/gpu/drm/xe/xe_guc_ads.c | 5 +++++ + drivers/gpu/drm/xe/xe_wa_oob.rules | 2 ++ + 3 files changed, 8 insertions(+) + +diff --git a/drivers/gpu/drm/xe/abi/guc_klvs_abi.h b/drivers/gpu/drm/xe/abi/guc_klvs_abi.h +index d633f1c739e43..7de8f827281fc 100644 +--- a/drivers/gpu/drm/xe/abi/guc_klvs_abi.h ++++ b/drivers/gpu/drm/xe/abi/guc_klvs_abi.h +@@ -367,6 +367,7 @@ enum xe_guc_klv_ids { + GUC_WA_KLV_NP_RD_WRITE_TO_CLEAR_RCSM_AT_CGP_LATE_RESTORE = 0x9008, + GUC_WORKAROUND_KLV_ID_BACK_TO_BACK_RCS_ENGINE_RESET = 0x9009, + GUC_WA_KLV_WAKE_POWER_DOMAINS_FOR_OUTBOUND_MMIO = 0x900a, ++ GUC_WA_KLV_RESET_BB_STACK_PTR_ON_VF_SWITCH = 0x900b, + }; + + #endif +diff --git a/drivers/gpu/drm/xe/xe_guc_ads.c b/drivers/gpu/drm/xe/xe_guc_ads.c +index 7031542a70ceb..1fccdef9e532e 100644 +--- a/drivers/gpu/drm/xe/xe_guc_ads.c ++++ b/drivers/gpu/drm/xe/xe_guc_ads.c +@@ -376,6 +376,11 @@ static void guc_waklv_init(struct xe_guc_ads *ads) + GUC_WORKAROUND_KLV_ID_BACK_TO_BACK_RCS_ENGINE_RESET, + &offset, &remain); + ++ if (GUC_FIRMWARE_VER(>->uc.guc) >= MAKE_GUC_VER(70, 44, 0) && XE_WA(gt, 16026508708)) ++ guc_waklv_enable_simple(ads, ++ GUC_WA_KLV_RESET_BB_STACK_PTR_ON_VF_SWITCH, ++ &offset, &remain); ++ + size = guc_ads_waklv_size(ads) - remain; + if (!size) + return; +diff --git a/drivers/gpu/drm/xe/xe_wa_oob.rules b/drivers/gpu/drm/xe/xe_wa_oob.rules +index 9b9e176992a83..9efc5accd43d1 100644 +--- a/drivers/gpu/drm/xe/xe_wa_oob.rules ++++ b/drivers/gpu/drm/xe/xe_wa_oob.rules +@@ -57,3 +57,5 @@ no_media_l3 MEDIA_VERSION(3000) + GRAPHICS_VERSION(1260), GRAPHICS_STEP(A0, B0) + 16023105232 GRAPHICS_VERSION_RANGE(2001, 3001) + MEDIA_VERSION_RANGE(1301, 3000) ++16026508708 GRAPHICS_VERSION_RANGE(1200, 3001) ++ MEDIA_VERSION_RANGE(1300, 3000) +-- +2.39.5 + diff --git a/queue-6.15/drm-xe-guc_pc-add-_locked-variant-for-min-max-freq.patch b/queue-6.15/drm-xe-guc_pc-add-_locked-variant-for-min-max-freq.patch new file mode 100644 index 0000000000..2972389183 --- /dev/null +++ b/queue-6.15/drm-xe-guc_pc-add-_locked-variant-for-min-max-freq.patch @@ -0,0 +1,223 @@ +From 443ff645b1edb328551eb979cfe5081ed36120bb Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Wed, 18 Jun 2025 11:49:58 -0700 +Subject: drm/xe/guc_pc: Add _locked variant for min/max freq + +From: Lucas De Marchi + +[ Upstream commit d8390768dcf6f5a78af56aa03797a076871b01f3 ] + +There are places in which the getters/setters are called one after the +other causing a multiple lock()/unlock(). These are not currently a +problem since they are all happening from the same thread, but there's a +race possibility as calls are added outside of the early init when the +max/min and stashed values need to be correlated. + +Add the _locked() variants to prepare for that. + +Reviewed-by: Rodrigo Vivi +Link: https://lore.kernel.org/r/20250618-wa-22019338487-v5-1-b888388477f2@intel.com +Signed-off-by: Lucas De Marchi +(cherry picked from commit 1beae9aa2b88d3a02eb666e7b777eb2d7bc645f4) +Signed-off-by: Lucas De Marchi +Stable-dep-of: 84c0b4a00610 ("drm/xe/bmg: Update Wa_22019338487") +Signed-off-by: Sasha Levin +--- + drivers/gpu/drm/xe/xe_guc_pc.c | 123 ++++++++++++++++++--------------- + 1 file changed, 69 insertions(+), 54 deletions(-) + +diff --git a/drivers/gpu/drm/xe/xe_guc_pc.c b/drivers/gpu/drm/xe/xe_guc_pc.c +index a7b8bacfe64ef..23a4c525c03bf 100644 +--- a/drivers/gpu/drm/xe/xe_guc_pc.c ++++ b/drivers/gpu/drm/xe/xe_guc_pc.c +@@ -5,6 +5,7 @@ + + #include "xe_guc_pc.h" + ++#include + #include + #include + +@@ -538,6 +539,25 @@ u32 xe_guc_pc_get_rpn_freq(struct xe_guc_pc *pc) + return pc->rpn_freq; + } + ++static int xe_guc_pc_get_min_freq_locked(struct xe_guc_pc *pc, u32 *freq) ++{ ++ int ret; ++ ++ lockdep_assert_held(&pc->freq_lock); ++ ++ /* Might be in the middle of a gt reset */ ++ if (!pc->freq_ready) ++ return -EAGAIN; ++ ++ ret = pc_action_query_task_state(pc); ++ if (ret) ++ return ret; ++ ++ *freq = pc_get_min_freq(pc); ++ ++ return 0; ++} ++ + /** + * xe_guc_pc_get_min_freq - Get the min operational frequency + * @pc: The GuC PC +@@ -547,27 +567,29 @@ u32 xe_guc_pc_get_rpn_freq(struct xe_guc_pc *pc) + * -EAGAIN if GuC PC not ready (likely in middle of a reset). + */ + int xe_guc_pc_get_min_freq(struct xe_guc_pc *pc, u32 *freq) ++{ ++ guard(mutex)(&pc->freq_lock); ++ ++ return xe_guc_pc_get_min_freq_locked(pc, freq); ++} ++ ++static int xe_guc_pc_set_min_freq_locked(struct xe_guc_pc *pc, u32 freq) + { + int ret; + +- xe_device_assert_mem_access(pc_to_xe(pc)); ++ lockdep_assert_held(&pc->freq_lock); + +- mutex_lock(&pc->freq_lock); +- if (!pc->freq_ready) { +- /* Might be in the middle of a gt reset */ +- ret = -EAGAIN; +- goto out; +- } ++ /* Might be in the middle of a gt reset */ ++ if (!pc->freq_ready) ++ return -EAGAIN; + +- ret = pc_action_query_task_state(pc); ++ ret = pc_set_min_freq(pc, freq); + if (ret) +- goto out; ++ return ret; + +- *freq = pc_get_min_freq(pc); ++ pc->user_requested_min = freq; + +-out: +- mutex_unlock(&pc->freq_lock); +- return ret; ++ return 0; + } + + /** +@@ -580,25 +602,29 @@ int xe_guc_pc_get_min_freq(struct xe_guc_pc *pc, u32 *freq) + * -EINVAL if value out of bounds. + */ + int xe_guc_pc_set_min_freq(struct xe_guc_pc *pc, u32 freq) ++{ ++ guard(mutex)(&pc->freq_lock); ++ ++ return xe_guc_pc_set_min_freq_locked(pc, freq); ++} ++ ++static int xe_guc_pc_get_max_freq_locked(struct xe_guc_pc *pc, u32 *freq) + { + int ret; + +- mutex_lock(&pc->freq_lock); +- if (!pc->freq_ready) { +- /* Might be in the middle of a gt reset */ +- ret = -EAGAIN; +- goto out; +- } ++ lockdep_assert_held(&pc->freq_lock); + +- ret = pc_set_min_freq(pc, freq); ++ /* Might be in the middle of a gt reset */ ++ if (!pc->freq_ready) ++ return -EAGAIN; ++ ++ ret = pc_action_query_task_state(pc); + if (ret) +- goto out; ++ return ret; + +- pc->user_requested_min = freq; ++ *freq = pc_get_max_freq(pc); + +-out: +- mutex_unlock(&pc->freq_lock); +- return ret; ++ return 0; + } + + /** +@@ -610,25 +636,29 @@ int xe_guc_pc_set_min_freq(struct xe_guc_pc *pc, u32 freq) + * -EAGAIN if GuC PC not ready (likely in middle of a reset). + */ + int xe_guc_pc_get_max_freq(struct xe_guc_pc *pc, u32 *freq) ++{ ++ guard(mutex)(&pc->freq_lock); ++ ++ return xe_guc_pc_get_max_freq_locked(pc, freq); ++} ++ ++static int xe_guc_pc_set_max_freq_locked(struct xe_guc_pc *pc, u32 freq) + { + int ret; + +- mutex_lock(&pc->freq_lock); +- if (!pc->freq_ready) { +- /* Might be in the middle of a gt reset */ +- ret = -EAGAIN; +- goto out; +- } ++ lockdep_assert_held(&pc->freq_lock); + +- ret = pc_action_query_task_state(pc); ++ /* Might be in the middle of a gt reset */ ++ if (!pc->freq_ready) ++ return -EAGAIN; ++ ++ ret = pc_set_max_freq(pc, freq); + if (ret) +- goto out; ++ return ret; + +- *freq = pc_get_max_freq(pc); ++ pc->user_requested_max = freq; + +-out: +- mutex_unlock(&pc->freq_lock); +- return ret; ++ return 0; + } + + /** +@@ -642,24 +672,9 @@ int xe_guc_pc_get_max_freq(struct xe_guc_pc *pc, u32 *freq) + */ + int xe_guc_pc_set_max_freq(struct xe_guc_pc *pc, u32 freq) + { +- int ret; +- +- mutex_lock(&pc->freq_lock); +- if (!pc->freq_ready) { +- /* Might be in the middle of a gt reset */ +- ret = -EAGAIN; +- goto out; +- } +- +- ret = pc_set_max_freq(pc, freq); +- if (ret) +- goto out; ++ guard(mutex)(&pc->freq_lock); + +- pc->user_requested_max = freq; +- +-out: +- mutex_unlock(&pc->freq_lock); +- return ret; ++ return xe_guc_pc_set_max_freq_locked(pc, freq); + } + + /** +-- +2.39.5 + diff --git a/queue-6.15/drm-xe-split-xe_device_td_flush.patch b/queue-6.15/drm-xe-split-xe_device_td_flush.patch new file mode 100644 index 0000000000..29bb83ce34 --- /dev/null +++ b/queue-6.15/drm-xe-split-xe_device_td_flush.patch @@ -0,0 +1,133 @@ +From 1124d79525844089be84655b448e9765372f34b4 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Wed, 18 Jun 2025 11:50:00 -0700 +Subject: drm/xe: Split xe_device_td_flush() + +From: Lucas De Marchi + +[ Upstream commit a1eec6cae95a1a0888cb8370338822ca81cd9436 ] + +xe_device_td_flush() has 2 possible implementations: an entire L2 flush +or a transient flush, depending on WA 16023588340. Make this clear by +splitting the function so it calls each of them. + +Reviewed-by: Matthew Auld +Link: https://lore.kernel.org/r/20250618-wa-22019338487-v5-3-b888388477f2@intel.com +Signed-off-by: Lucas De Marchi +(cherry picked from commit 5e300ed8a545bdffc26b579c526b5fef7b2d5365) +Signed-off-by: Lucas De Marchi +Stable-dep-of: 84c0b4a00610 ("drm/xe/bmg: Update Wa_22019338487") +Signed-off-by: Sasha Levin +--- + drivers/gpu/drm/xe/xe_device.c | 68 ++++++++++++++++++++-------------- + 1 file changed, 40 insertions(+), 28 deletions(-) + +diff --git a/drivers/gpu/drm/xe/xe_device.c b/drivers/gpu/drm/xe/xe_device.c +index 00191227bc95c..38fdddd7262aa 100644 +--- a/drivers/gpu/drm/xe/xe_device.c ++++ b/drivers/gpu/drm/xe/xe_device.c +@@ -972,38 +972,15 @@ void xe_device_wmb(struct xe_device *xe) + xe_mmio_write32(xe_root_tile_mmio(xe), VF_CAP_REG, 0); + } + +-/** +- * xe_device_td_flush() - Flush transient L3 cache entries +- * @xe: The device +- * +- * Display engine has direct access to memory and is never coherent with L3/L4 +- * caches (or CPU caches), however KMD is responsible for specifically flushing +- * transient L3 GPU cache entries prior to the flip sequence to ensure scanout +- * can happen from such a surface without seeing corruption. +- * +- * Display surfaces can be tagged as transient by mapping it using one of the +- * various L3:XD PAT index modes on Xe2. +- * +- * Note: On non-discrete xe2 platforms, like LNL, the entire L3 cache is flushed +- * at the end of each submission via PIPE_CONTROL for compute/render, since SA +- * Media is not coherent with L3 and we want to support render-vs-media +- * usescases. For other engines like copy/blt the HW internally forces uncached +- * behaviour, hence why we can skip the TDF on such platforms. ++/* ++ * Issue a TRANSIENT_FLUSH_REQUEST and wait for completion on each gt. + */ +-void xe_device_td_flush(struct xe_device *xe) ++static void tdf_request_sync(struct xe_device *xe) + { +- struct xe_gt *gt; + unsigned int fw_ref; ++ struct xe_gt *gt; + u8 id; + +- if (!IS_DGFX(xe) || GRAPHICS_VER(xe) < 20) +- return; +- +- if (XE_WA(xe_root_mmio_gt(xe), 16023588340)) { +- xe_device_l2_flush(xe); +- return; +- } +- + for_each_gt(gt, xe, id) { + if (xe_gt_is_media_type(gt)) + continue; +@@ -1013,6 +990,7 @@ void xe_device_td_flush(struct xe_device *xe) + return; + + xe_mmio_write32(>->mmio, XE2_TDF_CTRL, TRANSIENT_FLUSH_REQUEST); ++ + /* + * FIXME: We can likely do better here with our choice of + * timeout. Currently we just assume the worst case, i.e. 150us, +@@ -1043,15 +1021,49 @@ void xe_device_l2_flush(struct xe_device *xe) + return; + + spin_lock(>->global_invl_lock); +- xe_mmio_write32(>->mmio, XE2_GLOBAL_INVAL, 0x1); + ++ xe_mmio_write32(>->mmio, XE2_GLOBAL_INVAL, 0x1); + if (xe_mmio_wait32(>->mmio, XE2_GLOBAL_INVAL, 0x1, 0x0, 500, NULL, true)) + xe_gt_err_once(gt, "Global invalidation timeout\n"); ++ + spin_unlock(>->global_invl_lock); + + xe_force_wake_put(gt_to_fw(gt), fw_ref); + } + ++/** ++ * xe_device_td_flush() - Flush transient L3 cache entries ++ * @xe: The device ++ * ++ * Display engine has direct access to memory and is never coherent with L3/L4 ++ * caches (or CPU caches), however KMD is responsible for specifically flushing ++ * transient L3 GPU cache entries prior to the flip sequence to ensure scanout ++ * can happen from such a surface without seeing corruption. ++ * ++ * Display surfaces can be tagged as transient by mapping it using one of the ++ * various L3:XD PAT index modes on Xe2. ++ * ++ * Note: On non-discrete xe2 platforms, like LNL, the entire L3 cache is flushed ++ * at the end of each submission via PIPE_CONTROL for compute/render, since SA ++ * Media is not coherent with L3 and we want to support render-vs-media ++ * usescases. For other engines like copy/blt the HW internally forces uncached ++ * behaviour, hence why we can skip the TDF on such platforms. ++ */ ++void xe_device_td_flush(struct xe_device *xe) ++{ ++ struct xe_gt *root_gt; ++ ++ if (!IS_DGFX(xe) || GRAPHICS_VER(xe) < 20) ++ return; ++ ++ root_gt = xe_root_mmio_gt(xe); ++ if (XE_WA(root_gt, 16023588340)) ++ /* A transient flush is not sufficient: flush the L2 */ ++ xe_device_l2_flush(xe); ++ else ++ tdf_request_sync(xe); ++} ++ + u32 xe_device_ccs_bytes(struct xe_device *xe, u64 size) + { + return xe_device_has_flat_ccs(xe) ? +-- +2.39.5 + diff --git a/queue-6.15/enic-fix-incorrect-mtu-comparison-in-enic_change_mtu.patch b/queue-6.15/enic-fix-incorrect-mtu-comparison-in-enic_change_mtu.patch new file mode 100644 index 0000000000..632ccd5299 --- /dev/null +++ b/queue-6.15/enic-fix-incorrect-mtu-comparison-in-enic_change_mtu.patch @@ -0,0 +1,47 @@ +From 5eb5da69c1aafd2104085d577b09276f0bcfe034 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Sat, 28 Jun 2025 07:56:05 -0700 +Subject: enic: fix incorrect MTU comparison in enic_change_mtu() + +From: Alok Tiwari + +[ Upstream commit aaf2b2480375099c022a82023e1cd772bf1c6a5d ] + +The comparison in enic_change_mtu() incorrectly used the current +netdev->mtu instead of the new new_mtu value when warning about +an MTU exceeding the port MTU. This could suppress valid warnings +or issue incorrect ones. + +Fix the condition and log to properly reflect the new_mtu. + +Fixes: ab123fe071c9 ("enic: handle mtu change for vf properly") +Signed-off-by: Alok Tiwari +Acked-by: John Daley +Reviewed-by: Simon Horman +Link: https://patch.msgid.link/20250628145612.476096-1-alok.a.tiwari@oracle.com +Signed-off-by: Jakub Kicinski +Signed-off-by: Sasha Levin +--- + drivers/net/ethernet/cisco/enic/enic_main.c | 4 ++-- + 1 file changed, 2 insertions(+), 2 deletions(-) + +diff --git a/drivers/net/ethernet/cisco/enic/enic_main.c b/drivers/net/ethernet/cisco/enic/enic_main.c +index c753c35b26ebd..5a8ca5be9ca00 100644 +--- a/drivers/net/ethernet/cisco/enic/enic_main.c ++++ b/drivers/net/ethernet/cisco/enic/enic_main.c +@@ -1864,10 +1864,10 @@ static int enic_change_mtu(struct net_device *netdev, int new_mtu) + if (enic_is_dynamic(enic) || enic_is_sriov_vf(enic)) + return -EOPNOTSUPP; + +- if (netdev->mtu > enic->port_mtu) ++ if (new_mtu > enic->port_mtu) + netdev_warn(netdev, + "interface MTU (%d) set higher than port MTU (%d)\n", +- netdev->mtu, enic->port_mtu); ++ new_mtu, enic->port_mtu); + + return _enic_change_mtu(netdev, new_mtu); + } +-- +2.39.5 + diff --git a/queue-6.15/ethernet-atl1-add-missing-dma-mapping-error-checks-a.patch b/queue-6.15/ethernet-atl1-add-missing-dma-mapping-error-checks-a.patch new file mode 100644 index 0000000000..b84e9b494b --- /dev/null +++ b/queue-6.15/ethernet-atl1-add-missing-dma-mapping-error-checks-a.patch @@ -0,0 +1,213 @@ +From 6ee823e79d1d10ff48ac76d60ebf2621fd3d7e17 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Wed, 25 Jun 2025 16:16:24 +0200 +Subject: ethernet: atl1: Add missing DMA mapping error checks and count errors + +From: Thomas Fourier + +[ Upstream commit d72411d20905180cdc452c553be17481b24463d2 ] + +The `dma_map_XXX()` functions can fail and must be checked using +`dma_mapping_error()`. This patch adds proper error handling for all +DMA mapping calls. + +In `atl1_alloc_rx_buffers()`, if DMA mapping fails, the buffer is +deallocated and marked accordingly. + +In `atl1_tx_map()`, previously mapped buffers are unmapped and the +packet is dropped on failure. + +If `atl1_xmit_frame()` drops the packet, increment the tx_error counter. + +Fixes: f3cc28c79760 ("Add Attansic L1 ethernet driver.") +Signed-off-by: Thomas Fourier +Link: https://patch.msgid.link/20250625141629.114984-2-fourier.thomas@gmail.com +Signed-off-by: Jakub Kicinski +Signed-off-by: Sasha Levin +--- + drivers/net/ethernet/atheros/atlx/atl1.c | 79 +++++++++++++++++------- + 1 file changed, 57 insertions(+), 22 deletions(-) + +diff --git a/drivers/net/ethernet/atheros/atlx/atl1.c b/drivers/net/ethernet/atheros/atlx/atl1.c +index 38cd84b7677cf..86663aa0de17e 100644 +--- a/drivers/net/ethernet/atheros/atlx/atl1.c ++++ b/drivers/net/ethernet/atheros/atlx/atl1.c +@@ -1861,14 +1861,21 @@ static u16 atl1_alloc_rx_buffers(struct atl1_adapter *adapter) + break; + } + +- buffer_info->alloced = 1; +- buffer_info->skb = skb; +- buffer_info->length = (u16) adapter->rx_buffer_len; + page = virt_to_page(skb->data); + offset = offset_in_page(skb->data); + buffer_info->dma = dma_map_page(&pdev->dev, page, offset, + adapter->rx_buffer_len, + DMA_FROM_DEVICE); ++ if (dma_mapping_error(&pdev->dev, buffer_info->dma)) { ++ kfree_skb(skb); ++ adapter->soft_stats.rx_dropped++; ++ break; ++ } ++ ++ buffer_info->alloced = 1; ++ buffer_info->skb = skb; ++ buffer_info->length = (u16)adapter->rx_buffer_len; ++ + rfd_desc->buffer_addr = cpu_to_le64(buffer_info->dma); + rfd_desc->buf_len = cpu_to_le16(adapter->rx_buffer_len); + rfd_desc->coalese = 0; +@@ -2183,8 +2190,8 @@ static int atl1_tx_csum(struct atl1_adapter *adapter, struct sk_buff *skb, + return 0; + } + +-static void atl1_tx_map(struct atl1_adapter *adapter, struct sk_buff *skb, +- struct tx_packet_desc *ptpd) ++static bool atl1_tx_map(struct atl1_adapter *adapter, struct sk_buff *skb, ++ struct tx_packet_desc *ptpd) + { + struct atl1_tpd_ring *tpd_ring = &adapter->tpd_ring; + struct atl1_buffer *buffer_info; +@@ -2194,6 +2201,7 @@ static void atl1_tx_map(struct atl1_adapter *adapter, struct sk_buff *skb, + unsigned int nr_frags; + unsigned int f; + int retval; ++ u16 first_mapped; + u16 next_to_use; + u16 data_len; + u8 hdr_len; +@@ -2201,6 +2209,7 @@ static void atl1_tx_map(struct atl1_adapter *adapter, struct sk_buff *skb, + buf_len -= skb->data_len; + nr_frags = skb_shinfo(skb)->nr_frags; + next_to_use = atomic_read(&tpd_ring->next_to_use); ++ first_mapped = next_to_use; + buffer_info = &tpd_ring->buffer_info[next_to_use]; + BUG_ON(buffer_info->skb); + /* put skb in last TPD */ +@@ -2216,6 +2225,8 @@ static void atl1_tx_map(struct atl1_adapter *adapter, struct sk_buff *skb, + buffer_info->dma = dma_map_page(&adapter->pdev->dev, page, + offset, hdr_len, + DMA_TO_DEVICE); ++ if (dma_mapping_error(&adapter->pdev->dev, buffer_info->dma)) ++ goto dma_err; + + if (++next_to_use == tpd_ring->count) + next_to_use = 0; +@@ -2242,6 +2253,9 @@ static void atl1_tx_map(struct atl1_adapter *adapter, struct sk_buff *skb, + page, offset, + buffer_info->length, + DMA_TO_DEVICE); ++ if (dma_mapping_error(&adapter->pdev->dev, ++ buffer_info->dma)) ++ goto dma_err; + if (++next_to_use == tpd_ring->count) + next_to_use = 0; + } +@@ -2254,6 +2268,8 @@ static void atl1_tx_map(struct atl1_adapter *adapter, struct sk_buff *skb, + buffer_info->dma = dma_map_page(&adapter->pdev->dev, page, + offset, buf_len, + DMA_TO_DEVICE); ++ if (dma_mapping_error(&adapter->pdev->dev, buffer_info->dma)) ++ goto dma_err; + if (++next_to_use == tpd_ring->count) + next_to_use = 0; + } +@@ -2277,6 +2293,9 @@ static void atl1_tx_map(struct atl1_adapter *adapter, struct sk_buff *skb, + buffer_info->dma = skb_frag_dma_map(&adapter->pdev->dev, + frag, i * ATL1_MAX_TX_BUF_LEN, + buffer_info->length, DMA_TO_DEVICE); ++ if (dma_mapping_error(&adapter->pdev->dev, ++ buffer_info->dma)) ++ goto dma_err; + + if (++next_to_use == tpd_ring->count) + next_to_use = 0; +@@ -2285,6 +2304,22 @@ static void atl1_tx_map(struct atl1_adapter *adapter, struct sk_buff *skb, + + /* last tpd's buffer-info */ + buffer_info->skb = skb; ++ ++ return true; ++ ++ dma_err: ++ while (first_mapped != next_to_use) { ++ buffer_info = &tpd_ring->buffer_info[first_mapped]; ++ dma_unmap_page(&adapter->pdev->dev, ++ buffer_info->dma, ++ buffer_info->length, ++ DMA_TO_DEVICE); ++ buffer_info->dma = 0; ++ ++ if (++first_mapped == tpd_ring->count) ++ first_mapped = 0; ++ } ++ return false; + } + + static void atl1_tx_queue(struct atl1_adapter *adapter, u16 count, +@@ -2355,10 +2390,8 @@ static netdev_tx_t atl1_xmit_frame(struct sk_buff *skb, + + len = skb_headlen(skb); + +- if (unlikely(skb->len <= 0)) { +- dev_kfree_skb_any(skb); +- return NETDEV_TX_OK; +- } ++ if (unlikely(skb->len <= 0)) ++ goto drop_packet; + + nr_frags = skb_shinfo(skb)->nr_frags; + for (f = 0; f < nr_frags; f++) { +@@ -2371,10 +2404,9 @@ static netdev_tx_t atl1_xmit_frame(struct sk_buff *skb, + if (mss) { + if (skb->protocol == htons(ETH_P_IP)) { + proto_hdr_len = skb_tcp_all_headers(skb); +- if (unlikely(proto_hdr_len > len)) { +- dev_kfree_skb_any(skb); +- return NETDEV_TX_OK; +- } ++ if (unlikely(proto_hdr_len > len)) ++ goto drop_packet; ++ + /* need additional TPD ? */ + if (proto_hdr_len != len) + count += (len - proto_hdr_len + +@@ -2406,23 +2438,26 @@ static netdev_tx_t atl1_xmit_frame(struct sk_buff *skb, + } + + tso = atl1_tso(adapter, skb, ptpd); +- if (tso < 0) { +- dev_kfree_skb_any(skb); +- return NETDEV_TX_OK; +- } ++ if (tso < 0) ++ goto drop_packet; + + if (!tso) { + ret_val = atl1_tx_csum(adapter, skb, ptpd); +- if (ret_val < 0) { +- dev_kfree_skb_any(skb); +- return NETDEV_TX_OK; +- } ++ if (ret_val < 0) ++ goto drop_packet; + } + +- atl1_tx_map(adapter, skb, ptpd); ++ if (!atl1_tx_map(adapter, skb, ptpd)) ++ goto drop_packet; ++ + atl1_tx_queue(adapter, count, ptpd); + atl1_update_mailbox(adapter); + return NETDEV_TX_OK; ++ ++drop_packet: ++ adapter->soft_stats.tx_errors++; ++ dev_kfree_skb_any(skb); ++ return NETDEV_TX_OK; + } + + static int atl1_rings_clean(struct napi_struct *napi, int budget) +-- +2.39.5 + diff --git a/queue-6.15/firmware-arm_ffa-fix-memory-leak-by-freeing-notifier.patch b/queue-6.15/firmware-arm_ffa-fix-memory-leak-by-freeing-notifier.patch new file mode 100644 index 0000000000..654e33fa00 --- /dev/null +++ b/queue-6.15/firmware-arm_ffa-fix-memory-leak-by-freeing-notifier.patch @@ -0,0 +1,43 @@ +From ebb03bbd823c501d99857a7cb6e990b38ede8ec9 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Wed, 28 May 2025 09:49:41 +0100 +Subject: firmware: arm_ffa: Fix memory leak by freeing notifier callback node + +From: Sudeep Holla + +[ Upstream commit a833d31ad867103ba72a0b73f3606f4ab8601719 ] + +Commit e0573444edbf ("firmware: arm_ffa: Add interfaces to request +notification callbacks") adds support for notifier callbacks by allocating +and inserting a callback node into a hashtable during registration of +notifiers. However, during unregistration, the code only removes the +node from the hashtable without freeing the associated memory, resulting +in a memory leak. + +Resolve the memory leak issue by ensuring the allocated notifier callback +node is properly freed after it is removed from the hashtable entry. + +Fixes: e0573444edbf ("firmware: arm_ffa: Add interfaces to request notification callbacks") +Message-Id: <20250528-ffa_notif_fix-v1-1-5ed7bc7f8437@arm.com> +Reviewed-by: Jens Wiklander +Signed-off-by: Sudeep Holla +Signed-off-by: Sasha Levin +--- + drivers/firmware/arm_ffa/driver.c | 1 + + 1 file changed, 1 insertion(+) + +diff --git a/drivers/firmware/arm_ffa/driver.c b/drivers/firmware/arm_ffa/driver.c +index fe55613a8ea99..6f75cdf297209 100644 +--- a/drivers/firmware/arm_ffa/driver.c ++++ b/drivers/firmware/arm_ffa/driver.c +@@ -1284,6 +1284,7 @@ update_notifier_cb(struct ffa_device *dev, int notify_id, void *cb, + hash_add(drv_info->notifier_hash, &cb_info->hnode, notify_id); + } else { + hash_del(&cb_info->hnode); ++ kfree(cb_info); + } + + return 0; +-- +2.39.5 + diff --git a/queue-6.15/firmware-arm_ffa-fix-the-missing-entry-in-struct-ffa.patch b/queue-6.15/firmware-arm_ffa-fix-the-missing-entry-in-struct-ffa.patch new file mode 100644 index 0000000000..93c78208be --- /dev/null +++ b/queue-6.15/firmware-arm_ffa-fix-the-missing-entry-in-struct-ffa.patch @@ -0,0 +1,37 @@ +From bbe97f34e1a80f54642c235a556bd07844094ae0 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Tue, 3 Jun 2025 16:38:53 +0530 +Subject: firmware: arm_ffa: Fix the missing entry in struct + ffa_indirect_msg_hdr + +From: Viresh Kumar + +[ Upstream commit 4c46a471be12216347ba707f8eadadbf5d68e698 ] + +As per the spec, one 32 bit reserved entry is missing here, add it. + +Signed-off-by: Viresh Kumar +Fixes: 910cc1acc9b4 ("firmware: arm_ffa: Add support for passing UUID in FFA_MSG_SEND2") +Reviewed-by: Bertrand Marquis +Message-Id: <28a624fbf416975de4fbe08cfbf7c2db89cb630e.1748948911.git.viresh.kumar@linaro.org> +Signed-off-by: Sudeep Holla +Signed-off-by: Sasha Levin +--- + include/linux/arm_ffa.h | 1 + + 1 file changed, 1 insertion(+) + +diff --git a/include/linux/arm_ffa.h b/include/linux/arm_ffa.h +index 5bded24dc24fe..e1634897e159c 100644 +--- a/include/linux/arm_ffa.h ++++ b/include/linux/arm_ffa.h +@@ -283,6 +283,7 @@ struct ffa_indirect_msg_hdr { + u32 offset; + u32 send_recv_id; + u32 size; ++ u32 res1; + uuid_t uuid; + }; + +-- +2.39.5 + diff --git a/queue-6.15/firmware-arm_ffa-move-memory-allocation-outside-the-.patch b/queue-6.15/firmware-arm_ffa-move-memory-allocation-outside-the-.patch new file mode 100644 index 0000000000..22c69ce5ca --- /dev/null +++ b/queue-6.15/firmware-arm_ffa-move-memory-allocation-outside-the-.patch @@ -0,0 +1,135 @@ +From 7238f487414194e112dfc00d91e85067e555c467 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Wed, 28 May 2025 09:49:42 +0100 +Subject: firmware: arm_ffa: Move memory allocation outside the mutex locking + +From: Sudeep Holla + +[ Upstream commit 27e850c88df0e25474a8caeb2903e2e90b62c1dc ] + +The notifier callback node allocation is currently done while holding +the notify_lock mutex. While this is safe even if memory allocation may +sleep, we need to move the allocation outside the locked region in +preparation to move from using muxtes to rwlocks. + +Move the memory allocation to avoid potential sleeping in atomic context +once the locks are moved from mutex to rwlocks. + +Fixes: e0573444edbf ("firmware: arm_ffa: Add interfaces to request notification callbacks") +Message-Id: <20250528-ffa_notif_fix-v1-2-5ed7bc7f8437@arm.com> +Reviewed-by: Jens Wiklander +Signed-off-by: Sudeep Holla +Signed-off-by: Sasha Levin +--- + drivers/firmware/arm_ffa/driver.c | 48 +++++++++++++++---------------- + 1 file changed, 24 insertions(+), 24 deletions(-) + +diff --git a/drivers/firmware/arm_ffa/driver.c b/drivers/firmware/arm_ffa/driver.c +index 6f75cdf297209..44eecb786e67b 100644 +--- a/drivers/firmware/arm_ffa/driver.c ++++ b/drivers/firmware/arm_ffa/driver.c +@@ -1250,13 +1250,12 @@ notifier_hnode_get_by_type(u16 notify_id, enum notify_type type) + return NULL; + } + +-static int +-update_notifier_cb(struct ffa_device *dev, int notify_id, void *cb, +- void *cb_data, bool is_registration, bool is_framework) ++static int update_notifier_cb(struct ffa_device *dev, int notify_id, ++ struct notifier_cb_info *cb, bool is_framework) + { + struct notifier_cb_info *cb_info = NULL; + enum notify_type type = ffa_notify_type_get(dev->vm_id); +- bool cb_found; ++ bool cb_found, is_registration = !!cb; + + if (is_framework) + cb_info = notifier_hnode_get_by_vmid_uuid(notify_id, dev->vm_id, +@@ -1270,18 +1269,7 @@ update_notifier_cb(struct ffa_device *dev, int notify_id, void *cb, + return -EINVAL; + + if (is_registration) { +- cb_info = kzalloc(sizeof(*cb_info), GFP_KERNEL); +- if (!cb_info) +- return -ENOMEM; +- +- cb_info->dev = dev; +- cb_info->cb_data = cb_data; +- if (is_framework) +- cb_info->fwk_cb = cb; +- else +- cb_info->cb = cb; +- +- hash_add(drv_info->notifier_hash, &cb_info->hnode, notify_id); ++ hash_add(drv_info->notifier_hash, &cb->hnode, notify_id); + } else { + hash_del(&cb_info->hnode); + kfree(cb_info); +@@ -1303,8 +1291,7 @@ static int __ffa_notify_relinquish(struct ffa_device *dev, int notify_id, + + mutex_lock(&drv_info->notify_lock); + +- rc = update_notifier_cb(dev, notify_id, NULL, NULL, false, +- is_framework); ++ rc = update_notifier_cb(dev, notify_id, NULL, is_framework); + if (rc) { + pr_err("Could not unregister notification callback\n"); + mutex_unlock(&drv_info->notify_lock); +@@ -1335,6 +1322,7 @@ static int __ffa_notify_request(struct ffa_device *dev, bool is_per_vcpu, + { + int rc; + u32 flags = 0; ++ struct notifier_cb_info *cb_info = NULL; + + if (ffa_notifications_disabled()) + return -EOPNOTSUPP; +@@ -1342,6 +1330,17 @@ static int __ffa_notify_request(struct ffa_device *dev, bool is_per_vcpu, + if (notify_id >= FFA_MAX_NOTIFICATIONS) + return -EINVAL; + ++ cb_info = kzalloc(sizeof(*cb_info), GFP_KERNEL); ++ if (!cb_info) ++ return -ENOMEM; ++ ++ cb_info->dev = dev; ++ cb_info->cb_data = cb_data; ++ if (is_framework) ++ cb_info->fwk_cb = cb; ++ else ++ cb_info->cb = cb; ++ + mutex_lock(&drv_info->notify_lock); + + if (!is_framework) { +@@ -1349,21 +1348,22 @@ static int __ffa_notify_request(struct ffa_device *dev, bool is_per_vcpu, + flags = PER_VCPU_NOTIFICATION_FLAG; + + rc = ffa_notification_bind(dev->vm_id, BIT(notify_id), flags); +- if (rc) { +- mutex_unlock(&drv_info->notify_lock); +- return rc; +- } ++ if (rc) ++ goto out_unlock_free; + } + +- rc = update_notifier_cb(dev, notify_id, cb, cb_data, true, +- is_framework); ++ rc = update_notifier_cb(dev, notify_id, cb_info, is_framework); + if (rc) { + pr_err("Failed to register callback for %d - %d\n", + notify_id, rc); + if (!is_framework) + ffa_notification_unbind(dev->vm_id, BIT(notify_id)); + } ++ ++out_unlock_free: + mutex_unlock(&drv_info->notify_lock); ++ if (rc) ++ kfree(cb_info); + + return rc; + } +-- +2.39.5 + diff --git a/queue-6.15/firmware-arm_ffa-replace-mutex-with-rwlock-to-avoid-.patch b/queue-6.15/firmware-arm_ffa-replace-mutex-with-rwlock-to-avoid-.patch new file mode 100644 index 0000000000..c134a661a2 --- /dev/null +++ b/queue-6.15/firmware-arm_ffa-replace-mutex-with-rwlock-to-avoid-.patch @@ -0,0 +1,152 @@ +From 70f5446fd9f645f4729ad0253974f120fdf6d677 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Wed, 28 May 2025 09:49:43 +0100 +Subject: firmware: arm_ffa: Replace mutex with rwlock to avoid sleep in atomic + context +MIME-Version: 1.0 +Content-Type: text/plain; charset=UTF-8 +Content-Transfer-Encoding: 8bit + +From: Sudeep Holla + +[ Upstream commit 9ca7a421229bbdfbe2e1e628cff5cfa782720a10 ] + +The current use of a mutex to protect the notifier hashtable accesses +can lead to issues in the atomic context. It results in the below +kernel warnings: + + | BUG: sleeping function called from invalid context at kernel/locking/mutex.c:258 + | in_atomic(): 1, irqs_disabled(): 1, non_block: 0, pid: 9, name: kworker/0:0 + | preempt_count: 1, expected: 0 + | RCU nest depth: 0, expected: 0 + | CPU: 0 UID: 0 PID: 9 Comm: kworker/0:0 Not tainted 6.14.0 #4 + | Workqueue: ffa_pcpu_irq_notification notif_pcpu_irq_work_fn + | Call trace: + | show_stack+0x18/0x24 (C) + | dump_stack_lvl+0x78/0x90 + | dump_stack+0x18/0x24 + | __might_resched+0x114/0x170 + | __might_sleep+0x48/0x98 + | mutex_lock+0x24/0x80 + | handle_notif_callbacks+0x54/0xe0 + | notif_get_and_handle+0x40/0x88 + | generic_exec_single+0x80/0xc0 + | smp_call_function_single+0xfc/0x1a0 + | notif_pcpu_irq_work_fn+0x2c/0x38 + | process_one_work+0x14c/0x2b4 + | worker_thread+0x2e4/0x3e0 + | kthread+0x13c/0x210 + | ret_from_fork+0x10/0x20 + +To address this, replace the mutex with an rwlock to protect the notifier +hashtable accesses. This ensures that read-side locking does not sleep and +multiple readers can acquire the lock concurrently, avoiding unnecessary +contention and potential deadlocks. Writer access remains exclusive, +preserving correctness. + +This change resolves warnings from lockdep about potential sleep in +atomic context. + +Cc: Jens Wiklander +Reported-by: Jérôme Forissier +Closes: https://github.com/OP-TEE/optee_os/issues/7394 +Fixes: e0573444edbf ("firmware: arm_ffa: Add interfaces to request notification callbacks") +Message-Id: <20250528-ffa_notif_fix-v1-3-5ed7bc7f8437@arm.com> +Reviewed-by: Jens Wiklander +Tested-by: Jens Wiklander +Signed-off-by: Sudeep Holla +Signed-off-by: Sasha Levin +--- + drivers/firmware/arm_ffa/driver.c | 22 +++++++++++----------- + 1 file changed, 11 insertions(+), 11 deletions(-) + +diff --git a/drivers/firmware/arm_ffa/driver.c b/drivers/firmware/arm_ffa/driver.c +index 44eecb786e67b..37eb2e6c2f9f4 100644 +--- a/drivers/firmware/arm_ffa/driver.c ++++ b/drivers/firmware/arm_ffa/driver.c +@@ -110,7 +110,7 @@ struct ffa_drv_info { + struct work_struct sched_recv_irq_work; + struct xarray partition_info; + DECLARE_HASHTABLE(notifier_hash, ilog2(FFA_MAX_NOTIFICATIONS)); +- struct mutex notify_lock; /* lock to protect notifier hashtable */ ++ rwlock_t notify_lock; /* lock to protect notifier hashtable */ + }; + + static struct ffa_drv_info *drv_info; +@@ -1289,19 +1289,19 @@ static int __ffa_notify_relinquish(struct ffa_device *dev, int notify_id, + if (notify_id >= FFA_MAX_NOTIFICATIONS) + return -EINVAL; + +- mutex_lock(&drv_info->notify_lock); ++ write_lock(&drv_info->notify_lock); + + rc = update_notifier_cb(dev, notify_id, NULL, is_framework); + if (rc) { + pr_err("Could not unregister notification callback\n"); +- mutex_unlock(&drv_info->notify_lock); ++ write_unlock(&drv_info->notify_lock); + return rc; + } + + if (!is_framework) + rc = ffa_notification_unbind(dev->vm_id, BIT(notify_id)); + +- mutex_unlock(&drv_info->notify_lock); ++ write_unlock(&drv_info->notify_lock); + + return rc; + } +@@ -1341,7 +1341,7 @@ static int __ffa_notify_request(struct ffa_device *dev, bool is_per_vcpu, + else + cb_info->cb = cb; + +- mutex_lock(&drv_info->notify_lock); ++ write_lock(&drv_info->notify_lock); + + if (!is_framework) { + if (is_per_vcpu) +@@ -1361,7 +1361,7 @@ static int __ffa_notify_request(struct ffa_device *dev, bool is_per_vcpu, + } + + out_unlock_free: +- mutex_unlock(&drv_info->notify_lock); ++ write_unlock(&drv_info->notify_lock); + if (rc) + kfree(cb_info); + +@@ -1407,9 +1407,9 @@ static void handle_notif_callbacks(u64 bitmap, enum notify_type type) + if (!(bitmap & 1)) + continue; + +- mutex_lock(&drv_info->notify_lock); ++ read_lock(&drv_info->notify_lock); + cb_info = notifier_hnode_get_by_type(notify_id, type); +- mutex_unlock(&drv_info->notify_lock); ++ read_unlock(&drv_info->notify_lock); + + if (cb_info && cb_info->cb) + cb_info->cb(notify_id, cb_info->cb_data); +@@ -1447,9 +1447,9 @@ static void handle_fwk_notif_callbacks(u32 bitmap) + + ffa_rx_release(); + +- mutex_lock(&drv_info->notify_lock); ++ read_lock(&drv_info->notify_lock); + cb_info = notifier_hnode_get_by_vmid_uuid(notify_id, target, &uuid); +- mutex_unlock(&drv_info->notify_lock); ++ read_unlock(&drv_info->notify_lock); + + if (cb_info && cb_info->fwk_cb) + cb_info->fwk_cb(notify_id, cb_info->cb_data, buf); +@@ -1974,7 +1974,7 @@ static void ffa_notifications_setup(void) + goto cleanup; + + hash_init(drv_info->notifier_hash); +- mutex_init(&drv_info->notify_lock); ++ rwlock_init(&drv_info->notify_lock); + + drv_info->notif_enabled = true; + return; +-- +2.39.5 + diff --git a/queue-6.15/firmware-exynos-acpm-fix-timeouts-on-xfers-handling.patch b/queue-6.15/firmware-exynos-acpm-fix-timeouts-on-xfers-handling.patch new file mode 100644 index 0000000000..db1aa1ab28 --- /dev/null +++ b/queue-6.15/firmware-exynos-acpm-fix-timeouts-on-xfers-handling.patch @@ -0,0 +1,110 @@ +From 10dda582265f07f7852cf54eaf14e92cccc6f3d9 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Fri, 6 Jun 2025 10:45:37 +0000 +Subject: firmware: exynos-acpm: fix timeouts on xfers handling +MIME-Version: 1.0 +Content-Type: text/plain; charset=UTF-8 +Content-Transfer-Encoding: 8bit + +From: Tudor Ambarus + +[ Upstream commit 8d2c2fa2209e83d0eb10f7330d8a0bbdc1df32ff ] + +The mailbox framework has a single inflight request at a time. If +a request is sent while another is still active, it will be queued +to the mailbox core ring buffer. + +ACPM protocol did not serialize the calls to the mailbox subsystem so we +could start the timeout ticks in parallel for multiple requests, while +just one was being inflight. + +Consider a hypothetical case where the xfer timeout is 100ms and an ACPM +transaction takes 90ms: + | 0ms: Message #0 is queued in mailbox layer and sent out, then sits + | at acpm_dequeue_by_polling() with a timeout of 100ms + | 1ms: Message #1 is queued in mailbox layer but not sent out yet. + | Since send_message() doesn't block, it also sits at + | acpm_dequeue_by_polling() with a timeout of 100ms + | ... + | 90ms: Message #0 is completed, txdone is called and message #1 is sent + | 101ms: Message #1 times out since the count started at 1ms. Even though + | it has only been inflight for 11ms. + +Fix the problem by moving mbox_send_message() and mbox_client_txdone() +immediately after the message has been written to the TX queue and while +still keeping the ACPM TX queue lock. We thus tie together the TX write +with the doorbell ring and mark the TX as done after the doorbell has +been rung. This guarantees that the doorbell has been rang before +starting the timeout ticks. We should also see some performance +improvement as we no longer wait to receive a response before ringing +the doorbell for the next request, so the ACPM firmware shall be able to +drain faster the TX queue. Another benefit is that requests are no +longer able to ring the doorbell one for the other, so it eases +debugging. Finally, the mailbox software queue will always contain a +single doorbell request due to the serialization done at the ACPM TX +queue level. Protocols like ACPM, that handle their own hardware queues +need a passthrough mailbox API, where they are able to just ring the +doorbell or flip a bit directly into the mailbox controller. The mailbox +software queue mechanism, the locking done into the mailbox core is not +really needed, so hopefully this lays the foundation for a passthrough +mailbox API. + +Reported-by: André Draszik +Fixes: a88927b534ba ("firmware: add Exynos ACPM protocol driver") +Signed-off-by: Tudor Ambarus +Link: https://lore.kernel.org/r/20250606-acpm-timeout-v2-1-306b1aa07a6c@linaro.org +Signed-off-by: Krzysztof Kozlowski +Signed-off-by: Sasha Levin +--- + drivers/firmware/samsung/exynos-acpm.c | 25 +++++++++---------------- + 1 file changed, 9 insertions(+), 16 deletions(-) + +diff --git a/drivers/firmware/samsung/exynos-acpm.c b/drivers/firmware/samsung/exynos-acpm.c +index e80cb7a8da8f2..520a9fd3b0fd3 100644 +--- a/drivers/firmware/samsung/exynos-acpm.c ++++ b/drivers/firmware/samsung/exynos-acpm.c +@@ -430,6 +430,9 @@ int acpm_do_xfer(const struct acpm_handle *handle, const struct acpm_xfer *xfer) + return -EOPNOTSUPP; + } + ++ msg.chan_id = xfer->acpm_chan_id; ++ msg.chan_type = EXYNOS_MBOX_CHAN_TYPE_DOORBELL; ++ + scoped_guard(mutex, &achan->tx_lock) { + tx_front = readl(achan->tx.front); + idx = (tx_front + 1) % achan->qlen; +@@ -446,25 +449,15 @@ int acpm_do_xfer(const struct acpm_handle *handle, const struct acpm_xfer *xfer) + + /* Advance TX front. */ + writel(idx, achan->tx.front); +- } + +- msg.chan_id = xfer->acpm_chan_id; +- msg.chan_type = EXYNOS_MBOX_CHAN_TYPE_DOORBELL; +- ret = mbox_send_message(achan->chan, (void *)&msg); +- if (ret < 0) +- return ret; +- +- ret = acpm_wait_for_message_response(achan, xfer); ++ ret = mbox_send_message(achan->chan, (void *)&msg); ++ if (ret < 0) ++ return ret; + +- /* +- * NOTE: we might prefer not to need the mailbox ticker to manage the +- * transfer queueing since the protocol layer queues things by itself. +- * Unfortunately, we have to kick the mailbox framework after we have +- * received our message. +- */ +- mbox_client_txdone(achan->chan, ret); ++ mbox_client_txdone(achan->chan, 0); ++ } + +- return ret; ++ return acpm_wait_for_message_response(achan, xfer); + } + + /** +-- +2.39.5 + diff --git a/queue-6.15/flexfiles-pnfs-update-stats-on-nfs4err_delay-for-v4..patch b/queue-6.15/flexfiles-pnfs-update-stats-on-nfs4err_delay-for-v4..patch new file mode 100644 index 0000000000..7555005dcc --- /dev/null +++ b/queue-6.15/flexfiles-pnfs-update-stats-on-nfs4err_delay-for-v4..patch @@ -0,0 +1,36 @@ +From 4a6c9f2ec581e1dfad0139a66e81d42491d70510 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Thu, 15 May 2025 21:04:15 +0200 +Subject: flexfiles/pNFS: update stats on NFS4ERR_DELAY for v4.1 DSes + +From: Tigran Mkrtchyan + +[ Upstream commit e3e3775392f3f0f3e3044f8c162bf47858e01759 ] + +On NFS4ERR_DELAY nfs slient updates its stats, but misses for +flexfiles v4.1 DSes. + +Signed-off-by: Tigran Mkrtchyan +Signed-off-by: Anna Schumaker +Stable-dep-of: 38074de35b01 ("NFSv4/flexfiles: Fix handling of NFS level errors in I/O") +Signed-off-by: Sasha Levin +--- + fs/nfs/flexfilelayout/flexfilelayout.c | 2 ++ + 1 file changed, 2 insertions(+) + +diff --git a/fs/nfs/flexfilelayout/flexfilelayout.c b/fs/nfs/flexfilelayout/flexfilelayout.c +index e6909cafab686..df48074605967 100644 +--- a/fs/nfs/flexfilelayout/flexfilelayout.c ++++ b/fs/nfs/flexfilelayout/flexfilelayout.c +@@ -1129,6 +1129,8 @@ static int ff_layout_async_handle_error_v4(struct rpc_task *task, + nfs4_schedule_session_recovery(clp->cl_session, task->tk_status); + break; + case -NFS4ERR_DELAY: ++ nfs_inc_stats(lseg->pls_layout->plh_inode, NFSIOS_DELAY); ++ fallthrough; + case -NFS4ERR_GRACE: + rpc_delay(task, FF_LAYOUT_POLL_RETRY_MAX); + break; +-- +2.39.5 + diff --git a/queue-6.15/fs-export-anon_inode_make_secure_inode-and-fix-secre.patch b/queue-6.15/fs-export-anon_inode_make_secure_inode-and-fix-secre.patch new file mode 100644 index 0000000000..9ee9312e40 --- /dev/null +++ b/queue-6.15/fs-export-anon_inode_make_secure_inode-and-fix-secre.patch @@ -0,0 +1,130 @@ +From 5e73477cd349fd3ddf56f94558e6aa323ee0cb04 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Fri, 20 Jun 2025 07:03:30 +0000 +Subject: fs: export anon_inode_make_secure_inode() and fix secretmem LSM + bypass + +From: Shivank Garg + +[ Upstream commit cbe4134ea4bc493239786220bd69cb8a13493190 ] + +Export anon_inode_make_secure_inode() to allow KVM guest_memfd to create +anonymous inodes with proper security context. This replaces the current +pattern of calling alloc_anon_inode() followed by +inode_init_security_anon() for creating security context manually. + +This change also fixes a security regression in secretmem where the +S_PRIVATE flag was not cleared after alloc_anon_inode(), causing +LSM/SELinux checks to be bypassed for secretmem file descriptors. + +As guest_memfd currently resides in the KVM module, we need to export this +symbol for use outside the core kernel. In the future, guest_memfd might be +moved to core-mm, at which point the symbols no longer would have to be +exported. When/if that happens is still unclear. + +Fixes: 2bfe15c52612 ("mm: create security context for memfd_secret inodes") +Suggested-by: David Hildenbrand +Suggested-by: Mike Rapoport +Signed-off-by: Shivank Garg +Link: https://lore.kernel.org/20250620070328.803704-3-shivankg@amd.com +Acked-by: "Mike Rapoport (Microsoft)" +Signed-off-by: Christian Brauner +Signed-off-by: Sasha Levin +--- + fs/anon_inodes.c | 23 ++++++++++++++++++----- + include/linux/fs.h | 2 ++ + mm/secretmem.c | 9 +-------- + 3 files changed, 21 insertions(+), 13 deletions(-) + +diff --git a/fs/anon_inodes.c b/fs/anon_inodes.c +index e51e7d88980a2..1d847a939f29a 100644 +--- a/fs/anon_inodes.c ++++ b/fs/anon_inodes.c +@@ -98,14 +98,25 @@ static struct file_system_type anon_inode_fs_type = { + .kill_sb = kill_anon_super, + }; + +-static struct inode *anon_inode_make_secure_inode( +- const char *name, +- const struct inode *context_inode) ++/** ++ * anon_inode_make_secure_inode - allocate an anonymous inode with security context ++ * @sb: [in] Superblock to allocate from ++ * @name: [in] Name of the class of the newfile (e.g., "secretmem") ++ * @context_inode: ++ * [in] Optional parent inode for security inheritance ++ * ++ * The function ensures proper security initialization through the LSM hook ++ * security_inode_init_security_anon(). ++ * ++ * Return: Pointer to new inode on success, ERR_PTR on failure. ++ */ ++struct inode *anon_inode_make_secure_inode(struct super_block *sb, const char *name, ++ const struct inode *context_inode) + { + struct inode *inode; + int error; + +- inode = alloc_anon_inode(anon_inode_mnt->mnt_sb); ++ inode = alloc_anon_inode(sb); + if (IS_ERR(inode)) + return inode; + inode->i_flags &= ~S_PRIVATE; +@@ -118,6 +129,7 @@ static struct inode *anon_inode_make_secure_inode( + } + return inode; + } ++EXPORT_SYMBOL_GPL_FOR_MODULES(anon_inode_make_secure_inode, "kvm"); + + static struct file *__anon_inode_getfile(const char *name, + const struct file_operations *fops, +@@ -132,7 +144,8 @@ static struct file *__anon_inode_getfile(const char *name, + return ERR_PTR(-ENOENT); + + if (make_inode) { +- inode = anon_inode_make_secure_inode(name, context_inode); ++ inode = anon_inode_make_secure_inode(anon_inode_mnt->mnt_sb, ++ name, context_inode); + if (IS_ERR(inode)) { + file = ERR_CAST(inode); + goto err; +diff --git a/include/linux/fs.h b/include/linux/fs.h +index ef41a8213a6fa..1fe7bf10d442c 100644 +--- a/include/linux/fs.h ++++ b/include/linux/fs.h +@@ -3552,6 +3552,8 @@ extern int simple_write_begin(struct file *file, struct address_space *mapping, + extern const struct address_space_operations ram_aops; + extern int always_delete_dentry(const struct dentry *); + extern struct inode *alloc_anon_inode(struct super_block *); ++struct inode *anon_inode_make_secure_inode(struct super_block *sb, const char *name, ++ const struct inode *context_inode); + extern int simple_nosetlease(struct file *, int, struct file_lease **, void **); + extern const struct dentry_operations simple_dentry_operations; + +diff --git a/mm/secretmem.c b/mm/secretmem.c +index 1b0a214ee5580..4662f2510ae5f 100644 +--- a/mm/secretmem.c ++++ b/mm/secretmem.c +@@ -195,18 +195,11 @@ static struct file *secretmem_file_create(unsigned long flags) + struct file *file; + struct inode *inode; + const char *anon_name = "[secretmem]"; +- int err; + +- inode = alloc_anon_inode(secretmem_mnt->mnt_sb); ++ inode = anon_inode_make_secure_inode(secretmem_mnt->mnt_sb, anon_name, NULL); + if (IS_ERR(inode)) + return ERR_CAST(inode); + +- err = security_inode_init_security_anon(inode, &QSTR(anon_name), NULL); +- if (err) { +- file = ERR_PTR(err); +- goto err_free_inode; +- } +- + file = alloc_file_pseudo(inode, secretmem_mnt, "secretmem", + O_RDWR, &secretmem_fops); + if (IS_ERR(file)) +-- +2.39.5 + diff --git a/queue-6.15/genirq-irq_sim-initialize-work-context-pointers-prop.patch b/queue-6.15/genirq-irq_sim-initialize-work-context-pointers-prop.patch new file mode 100644 index 0000000000..d621e87ea3 --- /dev/null +++ b/queue-6.15/genirq-irq_sim-initialize-work-context-pointers-prop.patch @@ -0,0 +1,37 @@ +From 22bc2708b5fb54a85c01666a98fef27ca4b00e04 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Thu, 12 Jun 2025 21:48:27 +0900 +Subject: genirq/irq_sim: Initialize work context pointers properly + +From: Gyeyoung Baek + +[ Upstream commit 8a2277a3c9e4cc5398f80821afe7ecbe9bdf2819 ] + +Initialize `ops` member's pointers properly by using kzalloc() instead of +kmalloc() when allocating the simulation work context. Otherwise the +pointers contain random content leading to invalid dereferencing. + +Signed-off-by: Gyeyoung Baek +Signed-off-by: Thomas Gleixner +Link: https://lore.kernel.org/all/20250612124827.63259-1-gye976@gmail.com +Signed-off-by: Sasha Levin +--- + kernel/irq/irq_sim.c | 2 +- + 1 file changed, 1 insertion(+), 1 deletion(-) + +diff --git a/kernel/irq/irq_sim.c b/kernel/irq/irq_sim.c +index 1a3d483548e2f..ae4c9cbd1b4b9 100644 +--- a/kernel/irq/irq_sim.c ++++ b/kernel/irq/irq_sim.c +@@ -202,7 +202,7 @@ struct irq_domain *irq_domain_create_sim_full(struct fwnode_handle *fwnode, + void *data) + { + struct irq_sim_work_ctx *work_ctx __free(kfree) = +- kmalloc(sizeof(*work_ctx), GFP_KERNEL); ++ kzalloc(sizeof(*work_ctx), GFP_KERNEL); + + if (!work_ctx) + return ERR_PTR(-ENOMEM); +-- +2.39.5 + diff --git a/queue-6.15/ib-mlx5-fix-potential-deadlock-in-mr-deregistration.patch b/queue-6.15/ib-mlx5-fix-potential-deadlock-in-mr-deregistration.patch new file mode 100644 index 0000000000..91f1e8e698 --- /dev/null +++ b/queue-6.15/ib-mlx5-fix-potential-deadlock-in-mr-deregistration.patch @@ -0,0 +1,268 @@ +From c64c7d665911ae0e23b21f5c3f51719cec8b21e9 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Mon, 16 Jun 2025 11:14:09 +0300 +Subject: IB/mlx5: Fix potential deadlock in MR deregistration +MIME-Version: 1.0 +Content-Type: text/plain; charset=UTF-8 +Content-Transfer-Encoding: 8bit + +From: Or Har-Toov + +[ Upstream commit 2ed25aa7f7711f508b6120e336f05cd9d49943c0 ] + +The issue arises when kzalloc() is invoked while holding umem_mutex or +any other lock acquired under umem_mutex. This is problematic because +kzalloc() can trigger fs_reclaim_aqcuire(), which may, in turn, invoke +mmu_notifier_invalidate_range_start(). This function can lead to +mlx5_ib_invalidate_range(), which attempts to acquire umem_mutex again, +resulting in a deadlock. + +The problematic flow: + CPU0 | CPU1 +---------------------------------------|------------------------------------------------ +mlx5_ib_dereg_mr() | + → revoke_mr() | + → mutex_lock(&umem_odp->umem_mutex) | + | mlx5_mkey_cache_init() + | → mutex_lock(&dev->cache.rb_lock) + | → mlx5r_cache_create_ent_locked() + | → kzalloc(GFP_KERNEL) + | → fs_reclaim() + | → mmu_notifier_invalidate_range_start() + | → mlx5_ib_invalidate_range() + | → mutex_lock(&umem_odp->umem_mutex) + → cache_ent_find_and_store() | + → mutex_lock(&dev->cache.rb_lock) | + +Additionally, when kzalloc() is called from within +cache_ent_find_and_store(), we encounter the same deadlock due to +re-acquisition of umem_mutex. + +Solve by releasing umem_mutex in dereg_mr() after umr_revoke_mr() +and before acquiring rb_lock. This ensures that we don't hold +umem_mutex while performing memory allocations that could trigger +the reclaim path. + +This change prevents the deadlock by ensuring proper lock ordering and +avoiding holding locks during memory allocation operations that could +trigger the reclaim path. + +The following lockdep warning demonstrates the deadlock: + + python3/20557 is trying to acquire lock: + ffff888387542128 (&umem_odp->umem_mutex){+.+.}-{4:4}, at: + mlx5_ib_invalidate_range+0x5b/0x550 [mlx5_ib] + + but task is already holding lock: + ffffffff82f6b840 (mmu_notifier_invalidate_range_start){+.+.}-{0:0}, at: + unmap_vmas+0x7b/0x1a0 + + which lock already depends on the new lock. + + the existing dependency chain (in reverse order) is: + + -> #3 (mmu_notifier_invalidate_range_start){+.+.}-{0:0}: + fs_reclaim_acquire+0x60/0xd0 + mem_cgroup_css_alloc+0x6f/0x9b0 + cgroup_init_subsys+0xa4/0x240 + cgroup_init+0x1c8/0x510 + start_kernel+0x747/0x760 + x86_64_start_reservations+0x25/0x30 + x86_64_start_kernel+0x73/0x80 + common_startup_64+0x129/0x138 + + -> #2 (fs_reclaim){+.+.}-{0:0}: + fs_reclaim_acquire+0x91/0xd0 + __kmalloc_cache_noprof+0x4d/0x4c0 + mlx5r_cache_create_ent_locked+0x75/0x620 [mlx5_ib] + mlx5_mkey_cache_init+0x186/0x360 [mlx5_ib] + mlx5_ib_stage_post_ib_reg_umr_init+0x3c/0x60 [mlx5_ib] + __mlx5_ib_add+0x4b/0x190 [mlx5_ib] + mlx5r_probe+0xd9/0x320 [mlx5_ib] + auxiliary_bus_probe+0x42/0x70 + really_probe+0xdb/0x360 + __driver_probe_device+0x8f/0x130 + driver_probe_device+0x1f/0xb0 + __driver_attach+0xd4/0x1f0 + bus_for_each_dev+0x79/0xd0 + bus_add_driver+0xf0/0x200 + driver_register+0x6e/0xc0 + __auxiliary_driver_register+0x6a/0xc0 + do_one_initcall+0x5e/0x390 + do_init_module+0x88/0x240 + init_module_from_file+0x85/0xc0 + idempotent_init_module+0x104/0x300 + __x64_sys_finit_module+0x68/0xc0 + do_syscall_64+0x6d/0x140 + entry_SYSCALL_64_after_hwframe+0x4b/0x53 + + -> #1 (&dev->cache.rb_lock){+.+.}-{4:4}: + __mutex_lock+0x98/0xf10 + __mlx5_ib_dereg_mr+0x6f2/0x890 [mlx5_ib] + mlx5_ib_dereg_mr+0x21/0x110 [mlx5_ib] + ib_dereg_mr_user+0x85/0x1f0 [ib_core] + uverbs_free_mr+0x19/0x30 [ib_uverbs] + destroy_hw_idr_uobject+0x21/0x80 [ib_uverbs] + uverbs_destroy_uobject+0x60/0x3d0 [ib_uverbs] + uobj_destroy+0x57/0xa0 [ib_uverbs] + ib_uverbs_cmd_verbs+0x4d5/0x1210 [ib_uverbs] + ib_uverbs_ioctl+0x129/0x230 [ib_uverbs] + __x64_sys_ioctl+0x596/0xaa0 + do_syscall_64+0x6d/0x140 + entry_SYSCALL_64_after_hwframe+0x4b/0x53 + + -> #0 (&umem_odp->umem_mutex){+.+.}-{4:4}: + __lock_acquire+0x1826/0x2f00 + lock_acquire+0xd3/0x2e0 + __mutex_lock+0x98/0xf10 + mlx5_ib_invalidate_range+0x5b/0x550 [mlx5_ib] + __mmu_notifier_invalidate_range_start+0x18e/0x1f0 + unmap_vmas+0x182/0x1a0 + exit_mmap+0xf3/0x4a0 + mmput+0x3a/0x100 + do_exit+0x2b9/0xa90 + do_group_exit+0x32/0xa0 + get_signal+0xc32/0xcb0 + arch_do_signal_or_restart+0x29/0x1d0 + syscall_exit_to_user_mode+0x105/0x1d0 + do_syscall_64+0x79/0x140 + entry_SYSCALL_64_after_hwframe+0x4b/0x53 + + Chain exists of: + &dev->cache.rb_lock --> mmu_notifier_invalidate_range_start --> + &umem_odp->umem_mutex + + Possible unsafe locking scenario: + + CPU0 CPU1 + ---- ---- + lock(&umem_odp->umem_mutex); + lock(mmu_notifier_invalidate_range_start); + lock(&umem_odp->umem_mutex); + lock(&dev->cache.rb_lock); + + *** DEADLOCK *** + +Fixes: abb604a1a9c8 ("RDMA/mlx5: Fix a race for an ODP MR which leads to CQE with error") +Signed-off-by: Or Har-Toov +Reviewed-by: Michael Guralnik +Link: https://patch.msgid.link/3c8f225a8a9fade647d19b014df1172544643e4a.1750061612.git.leon@kernel.org +Signed-off-by: Leon Romanovsky +Signed-off-by: Sasha Levin +--- + drivers/infiniband/hw/mlx5/mr.c | 61 +++++++++++++++++++++++++-------- + 1 file changed, 47 insertions(+), 14 deletions(-) + +diff --git a/drivers/infiniband/hw/mlx5/mr.c b/drivers/infiniband/hw/mlx5/mr.c +index 5fbebafc87742..cb403134eeaea 100644 +--- a/drivers/infiniband/hw/mlx5/mr.c ++++ b/drivers/infiniband/hw/mlx5/mr.c +@@ -2027,23 +2027,50 @@ void mlx5_ib_revoke_data_direct_mrs(struct mlx5_ib_dev *dev) + } + } + +-static int mlx5_revoke_mr(struct mlx5_ib_mr *mr) ++static int mlx5_umr_revoke_mr_with_lock(struct mlx5_ib_mr *mr) + { +- struct mlx5_ib_dev *dev = to_mdev(mr->ibmr.device); +- struct mlx5_cache_ent *ent = mr->mmkey.cache_ent; +- bool is_odp = is_odp_mr(mr); + bool is_odp_dma_buf = is_dmabuf_mr(mr) && +- !to_ib_umem_dmabuf(mr->umem)->pinned; +- bool from_cache = !!ent; +- int ret = 0; ++ !to_ib_umem_dmabuf(mr->umem)->pinned; ++ bool is_odp = is_odp_mr(mr); ++ int ret; + + if (is_odp) + mutex_lock(&to_ib_umem_odp(mr->umem)->umem_mutex); + + if (is_odp_dma_buf) +- dma_resv_lock(to_ib_umem_dmabuf(mr->umem)->attach->dmabuf->resv, NULL); ++ dma_resv_lock(to_ib_umem_dmabuf(mr->umem)->attach->dmabuf->resv, ++ NULL); ++ ++ ret = mlx5r_umr_revoke_mr(mr); ++ ++ if (is_odp) { ++ if (!ret) ++ to_ib_umem_odp(mr->umem)->private = NULL; ++ mutex_unlock(&to_ib_umem_odp(mr->umem)->umem_mutex); ++ } ++ ++ if (is_odp_dma_buf) { ++ if (!ret) ++ to_ib_umem_dmabuf(mr->umem)->private = NULL; ++ dma_resv_unlock( ++ to_ib_umem_dmabuf(mr->umem)->attach->dmabuf->resv); ++ } + +- if (mr->mmkey.cacheable && !mlx5r_umr_revoke_mr(mr) && !cache_ent_find_and_store(dev, mr)) { ++ return ret; ++} ++ ++static int mlx5r_handle_mkey_cleanup(struct mlx5_ib_mr *mr) ++{ ++ bool is_odp_dma_buf = is_dmabuf_mr(mr) && ++ !to_ib_umem_dmabuf(mr->umem)->pinned; ++ struct mlx5_ib_dev *dev = to_mdev(mr->ibmr.device); ++ struct mlx5_cache_ent *ent = mr->mmkey.cache_ent; ++ bool is_odp = is_odp_mr(mr); ++ bool from_cache = !!ent; ++ int ret; ++ ++ if (mr->mmkey.cacheable && !mlx5_umr_revoke_mr_with_lock(mr) && ++ !cache_ent_find_and_store(dev, mr)) { + ent = mr->mmkey.cache_ent; + /* upon storing to a clean temp entry - schedule its cleanup */ + spin_lock_irq(&ent->mkeys_queue.lock); +@@ -2055,7 +2082,7 @@ static int mlx5_revoke_mr(struct mlx5_ib_mr *mr) + ent->tmp_cleanup_scheduled = true; + } + spin_unlock_irq(&ent->mkeys_queue.lock); +- goto out; ++ return 0; + } + + if (ent) { +@@ -2064,8 +2091,14 @@ static int mlx5_revoke_mr(struct mlx5_ib_mr *mr) + mr->mmkey.cache_ent = NULL; + spin_unlock_irq(&ent->mkeys_queue.lock); + } ++ ++ if (is_odp) ++ mutex_lock(&to_ib_umem_odp(mr->umem)->umem_mutex); ++ ++ if (is_odp_dma_buf) ++ dma_resv_lock(to_ib_umem_dmabuf(mr->umem)->attach->dmabuf->resv, ++ NULL); + ret = destroy_mkey(dev, mr); +-out: + if (is_odp) { + if (!ret) + to_ib_umem_odp(mr->umem)->private = NULL; +@@ -2075,9 +2108,9 @@ static int mlx5_revoke_mr(struct mlx5_ib_mr *mr) + if (is_odp_dma_buf) { + if (!ret) + to_ib_umem_dmabuf(mr->umem)->private = NULL; +- dma_resv_unlock(to_ib_umem_dmabuf(mr->umem)->attach->dmabuf->resv); ++ dma_resv_unlock( ++ to_ib_umem_dmabuf(mr->umem)->attach->dmabuf->resv); + } +- + return ret; + } + +@@ -2126,7 +2159,7 @@ static int __mlx5_ib_dereg_mr(struct ib_mr *ibmr) + } + + /* Stop DMA */ +- rc = mlx5_revoke_mr(mr); ++ rc = mlx5r_handle_mkey_cleanup(mr); + if (rc) + return rc; + +-- +2.39.5 + diff --git a/queue-6.15/idpf-convert-control-queue-mutex-to-a-spinlock.patch b/queue-6.15/idpf-convert-control-queue-mutex-to-a-spinlock.patch new file mode 100644 index 0000000000..7ee42bf64e --- /dev/null +++ b/queue-6.15/idpf-convert-control-queue-mutex-to-a-spinlock.patch @@ -0,0 +1,232 @@ +From c54d7701d9e5b78a8eedfb64dab935314c6b5e51 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Fri, 23 May 2025 14:55:37 -0600 +Subject: idpf: convert control queue mutex to a spinlock + +From: Ahmed Zaki + +[ Upstream commit b2beb5bb2cd90d7939e470ed4da468683f41baa3 ] + +With VIRTCHNL2_CAP_MACFILTER enabled, the following warning is generated +on module load: + +[ 324.701677] BUG: sleeping function called from invalid context at kernel/locking/mutex.c:578 +[ 324.701684] in_atomic(): 1, irqs_disabled(): 0, non_block: 0, pid: 1582, name: NetworkManager +[ 324.701689] preempt_count: 201, expected: 0 +[ 324.701693] RCU nest depth: 0, expected: 0 +[ 324.701697] 2 locks held by NetworkManager/1582: +[ 324.701702] #0: ffffffff9f7be770 (rtnl_mutex){....}-{3:3}, at: rtnl_newlink+0x791/0x21e0 +[ 324.701730] #1: ff1100216c380368 (_xmit_ETHER){....}-{2:2}, at: __dev_open+0x3f0/0x870 +[ 324.701749] Preemption disabled at: +[ 324.701752] [] __dev_open+0x3dd/0x870 +[ 324.701765] CPU: 30 UID: 0 PID: 1582 Comm: NetworkManager Not tainted 6.15.0-rc5+ #2 PREEMPT(voluntary) +[ 324.701771] Hardware name: Intel Corporation M50FCP2SBSTD/M50FCP2SBSTD, BIOS SE5C741.86B.01.01.0001.2211140926 11/14/2022 +[ 324.701774] Call Trace: +[ 324.701777] +[ 324.701779] dump_stack_lvl+0x5d/0x80 +[ 324.701788] ? __dev_open+0x3dd/0x870 +[ 324.701793] __might_resched.cold+0x1ef/0x23d +<..> +[ 324.701818] __mutex_lock+0x113/0x1b80 +<..> +[ 324.701917] idpf_ctlq_clean_sq+0xad/0x4b0 [idpf] +[ 324.701935] ? kasan_save_track+0x14/0x30 +[ 324.701941] idpf_mb_clean+0x143/0x380 [idpf] +<..> +[ 324.701991] idpf_send_mb_msg+0x111/0x720 [idpf] +[ 324.702009] idpf_vc_xn_exec+0x4cc/0x990 [idpf] +[ 324.702021] ? rcu_is_watching+0x12/0xc0 +[ 324.702035] idpf_add_del_mac_filters+0x3ed/0xb50 [idpf] +<..> +[ 324.702122] __hw_addr_sync_dev+0x1cf/0x300 +[ 324.702126] ? find_held_lock+0x32/0x90 +[ 324.702134] idpf_set_rx_mode+0x317/0x390 [idpf] +[ 324.702152] __dev_open+0x3f8/0x870 +[ 324.702159] ? __pfx___dev_open+0x10/0x10 +[ 324.702174] __dev_change_flags+0x443/0x650 +<..> +[ 324.702208] netif_change_flags+0x80/0x160 +[ 324.702218] do_setlink.isra.0+0x16a0/0x3960 +<..> +[ 324.702349] rtnl_newlink+0x12fd/0x21e0 + +The sequence is as follows: + rtnl_newlink()-> + __dev_change_flags()-> + __dev_open()-> + dev_set_rx_mode() - > # disables BH and grabs "dev->addr_list_lock" + idpf_set_rx_mode() -> # proceed only if VIRTCHNL2_CAP_MACFILTER is ON + __dev_uc_sync() -> + idpf_add_mac_filter -> + idpf_add_del_mac_filters -> + idpf_send_mb_msg() -> + idpf_mb_clean() -> + idpf_ctlq_clean_sq() # mutex_lock(cq_lock) + +Fix by converting cq_lock to a spinlock. All operations under the new +lock are safe except freeing the DMA memory, which may use vunmap(). Fix +by requesting a contiguous physical memory for the DMA mapping. + +Fixes: a251eee62133 ("idpf: add SRIOV support and other ndo_ops") +Reviewed-by: Aleksandr Loktionov +Signed-off-by: Ahmed Zaki +Reviewed-by: Simon Horman +Tested-by: Samuel Salin +Signed-off-by: Tony Nguyen +Signed-off-by: Sasha Levin +--- + .../net/ethernet/intel/idpf/idpf_controlq.c | 23 +++++++++---------- + .../ethernet/intel/idpf/idpf_controlq_api.h | 2 +- + drivers/net/ethernet/intel/idpf/idpf_lib.c | 12 ++++++---- + 3 files changed, 20 insertions(+), 17 deletions(-) + +diff --git a/drivers/net/ethernet/intel/idpf/idpf_controlq.c b/drivers/net/ethernet/intel/idpf/idpf_controlq.c +index b28991dd18703..48b8e184f3db6 100644 +--- a/drivers/net/ethernet/intel/idpf/idpf_controlq.c ++++ b/drivers/net/ethernet/intel/idpf/idpf_controlq.c +@@ -96,7 +96,7 @@ static void idpf_ctlq_init_rxq_bufs(struct idpf_ctlq_info *cq) + */ + static void idpf_ctlq_shutdown(struct idpf_hw *hw, struct idpf_ctlq_info *cq) + { +- mutex_lock(&cq->cq_lock); ++ spin_lock(&cq->cq_lock); + + /* free ring buffers and the ring itself */ + idpf_ctlq_dealloc_ring_res(hw, cq); +@@ -104,8 +104,7 @@ static void idpf_ctlq_shutdown(struct idpf_hw *hw, struct idpf_ctlq_info *cq) + /* Set ring_size to 0 to indicate uninitialized queue */ + cq->ring_size = 0; + +- mutex_unlock(&cq->cq_lock); +- mutex_destroy(&cq->cq_lock); ++ spin_unlock(&cq->cq_lock); + } + + /** +@@ -173,7 +172,7 @@ int idpf_ctlq_add(struct idpf_hw *hw, + + idpf_ctlq_init_regs(hw, cq, is_rxq); + +- mutex_init(&cq->cq_lock); ++ spin_lock_init(&cq->cq_lock); + + list_add(&cq->cq_list, &hw->cq_list_head); + +@@ -272,7 +271,7 @@ int idpf_ctlq_send(struct idpf_hw *hw, struct idpf_ctlq_info *cq, + int err = 0; + int i; + +- mutex_lock(&cq->cq_lock); ++ spin_lock(&cq->cq_lock); + + /* Ensure there are enough descriptors to send all messages */ + num_desc_avail = IDPF_CTLQ_DESC_UNUSED(cq); +@@ -332,7 +331,7 @@ int idpf_ctlq_send(struct idpf_hw *hw, struct idpf_ctlq_info *cq, + wr32(hw, cq->reg.tail, cq->next_to_use); + + err_unlock: +- mutex_unlock(&cq->cq_lock); ++ spin_unlock(&cq->cq_lock); + + return err; + } +@@ -364,7 +363,7 @@ int idpf_ctlq_clean_sq(struct idpf_ctlq_info *cq, u16 *clean_count, + if (*clean_count > cq->ring_size) + return -EBADR; + +- mutex_lock(&cq->cq_lock); ++ spin_lock(&cq->cq_lock); + + ntc = cq->next_to_clean; + +@@ -397,7 +396,7 @@ int idpf_ctlq_clean_sq(struct idpf_ctlq_info *cq, u16 *clean_count, + + cq->next_to_clean = ntc; + +- mutex_unlock(&cq->cq_lock); ++ spin_unlock(&cq->cq_lock); + + /* Return number of descriptors actually cleaned */ + *clean_count = i; +@@ -435,7 +434,7 @@ int idpf_ctlq_post_rx_buffs(struct idpf_hw *hw, struct idpf_ctlq_info *cq, + if (*buff_count > 0) + buffs_avail = true; + +- mutex_lock(&cq->cq_lock); ++ spin_lock(&cq->cq_lock); + + if (tbp >= cq->ring_size) + tbp = 0; +@@ -524,7 +523,7 @@ int idpf_ctlq_post_rx_buffs(struct idpf_hw *hw, struct idpf_ctlq_info *cq, + wr32(hw, cq->reg.tail, cq->next_to_post); + } + +- mutex_unlock(&cq->cq_lock); ++ spin_unlock(&cq->cq_lock); + + /* return the number of buffers that were not posted */ + *buff_count = *buff_count - i; +@@ -552,7 +551,7 @@ int idpf_ctlq_recv(struct idpf_ctlq_info *cq, u16 *num_q_msg, + u16 i; + + /* take the lock before we start messing with the ring */ +- mutex_lock(&cq->cq_lock); ++ spin_lock(&cq->cq_lock); + + ntc = cq->next_to_clean; + +@@ -614,7 +613,7 @@ int idpf_ctlq_recv(struct idpf_ctlq_info *cq, u16 *num_q_msg, + + cq->next_to_clean = ntc; + +- mutex_unlock(&cq->cq_lock); ++ spin_unlock(&cq->cq_lock); + + *num_q_msg = i; + if (*num_q_msg == 0) +diff --git a/drivers/net/ethernet/intel/idpf/idpf_controlq_api.h b/drivers/net/ethernet/intel/idpf/idpf_controlq_api.h +index e8e046ef2f0d7..5890d8adca4a8 100644 +--- a/drivers/net/ethernet/intel/idpf/idpf_controlq_api.h ++++ b/drivers/net/ethernet/intel/idpf/idpf_controlq_api.h +@@ -99,7 +99,7 @@ struct idpf_ctlq_info { + + enum idpf_ctlq_type cq_type; + int q_id; +- struct mutex cq_lock; /* control queue lock */ ++ spinlock_t cq_lock; /* control queue lock */ + /* used for interrupt processing */ + u16 next_to_use; + u16 next_to_clean; +diff --git a/drivers/net/ethernet/intel/idpf/idpf_lib.c b/drivers/net/ethernet/intel/idpf/idpf_lib.c +index 2ed801398971c..fe96e20573660 100644 +--- a/drivers/net/ethernet/intel/idpf/idpf_lib.c ++++ b/drivers/net/ethernet/intel/idpf/idpf_lib.c +@@ -2329,8 +2329,12 @@ void *idpf_alloc_dma_mem(struct idpf_hw *hw, struct idpf_dma_mem *mem, u64 size) + struct idpf_adapter *adapter = hw->back; + size_t sz = ALIGN(size, 4096); + +- mem->va = dma_alloc_coherent(&adapter->pdev->dev, sz, +- &mem->pa, GFP_KERNEL); ++ /* The control queue resources are freed under a spinlock, contiguous ++ * pages will avoid IOMMU remapping and the use vmap (and vunmap in ++ * dma_free_*() path. ++ */ ++ mem->va = dma_alloc_attrs(&adapter->pdev->dev, sz, &mem->pa, ++ GFP_KERNEL, DMA_ATTR_FORCE_CONTIGUOUS); + mem->size = sz; + + return mem->va; +@@ -2345,8 +2349,8 @@ void idpf_free_dma_mem(struct idpf_hw *hw, struct idpf_dma_mem *mem) + { + struct idpf_adapter *adapter = hw->back; + +- dma_free_coherent(&adapter->pdev->dev, mem->size, +- mem->va, mem->pa); ++ dma_free_attrs(&adapter->pdev->dev, mem->size, ++ mem->va, mem->pa, DMA_ATTR_FORCE_CONTIGUOUS); + mem->size = 0; + mem->va = NULL; + mem->pa = 0; +-- +2.39.5 + diff --git a/queue-6.15/idpf-return-0-size-for-rss-key-if-not-supported.patch b/queue-6.15/idpf-return-0-size-for-rss-key-if-not-supported.patch new file mode 100644 index 0000000000..b01ffedb21 --- /dev/null +++ b/queue-6.15/idpf-return-0-size-for-rss-key-if-not-supported.patch @@ -0,0 +1,101 @@ +From 521f9753b3ac2ea6a221e5c3e9afde6c68d7a1a4 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Thu, 22 May 2025 10:52:06 +0200 +Subject: idpf: return 0 size for RSS key if not supported + +From: Michal Swiatkowski + +[ Upstream commit f77bf1ebf8ff6301ccdbc346f7b52db928f9cbf8 ] + +Returning -EOPNOTSUPP from function returning u32 is leading to +cast and invalid size value as a result. + +-EOPNOTSUPP as a size probably will lead to allocation fail. + +Command: ethtool -x eth0 +It is visible on all devices that don't have RSS caps set. + +[ 136.615917] Call Trace: +[ 136.615921] +[ 136.615927] ? __warn+0x89/0x130 +[ 136.615942] ? __alloc_frozen_pages_noprof+0x322/0x330 +[ 136.615953] ? report_bug+0x164/0x190 +[ 136.615968] ? handle_bug+0x58/0x90 +[ 136.615979] ? exc_invalid_op+0x17/0x70 +[ 136.615987] ? asm_exc_invalid_op+0x1a/0x20 +[ 136.616001] ? rss_prepare_get.constprop.0+0xb9/0x170 +[ 136.616016] ? __alloc_frozen_pages_noprof+0x322/0x330 +[ 136.616028] __alloc_pages_noprof+0xe/0x20 +[ 136.616038] ___kmalloc_large_node+0x80/0x110 +[ 136.616072] __kmalloc_large_node_noprof+0x1d/0xa0 +[ 136.616081] __kmalloc_noprof+0x32c/0x4c0 +[ 136.616098] ? rss_prepare_get.constprop.0+0xb9/0x170 +[ 136.616105] rss_prepare_get.constprop.0+0xb9/0x170 +[ 136.616114] ethnl_default_doit+0x107/0x3d0 +[ 136.616131] genl_family_rcv_msg_doit+0x100/0x160 +[ 136.616147] genl_rcv_msg+0x1b8/0x2c0 +[ 136.616156] ? __pfx_ethnl_default_doit+0x10/0x10 +[ 136.616168] ? __pfx_genl_rcv_msg+0x10/0x10 +[ 136.616176] netlink_rcv_skb+0x58/0x110 +[ 136.616186] genl_rcv+0x28/0x40 +[ 136.616195] netlink_unicast+0x19b/0x290 +[ 136.616206] netlink_sendmsg+0x222/0x490 +[ 136.616215] __sys_sendto+0x1fd/0x210 +[ 136.616233] __x64_sys_sendto+0x24/0x30 +[ 136.616242] do_syscall_64+0x82/0x160 +[ 136.616252] ? __sys_recvmsg+0x83/0xe0 +[ 136.616265] ? syscall_exit_to_user_mode+0x10/0x210 +[ 136.616275] ? do_syscall_64+0x8e/0x160 +[ 136.616282] ? __count_memcg_events+0xa1/0x130 +[ 136.616295] ? count_memcg_events.constprop.0+0x1a/0x30 +[ 136.616306] ? handle_mm_fault+0xae/0x2d0 +[ 136.616319] ? do_user_addr_fault+0x379/0x670 +[ 136.616328] ? clear_bhb_loop+0x45/0xa0 +[ 136.616340] ? clear_bhb_loop+0x45/0xa0 +[ 136.616349] ? clear_bhb_loop+0x45/0xa0 +[ 136.616359] entry_SYSCALL_64_after_hwframe+0x76/0x7e +[ 136.616369] RIP: 0033:0x7fd30ba7b047 +[ 136.616376] Code: 0c 00 f7 d8 64 89 02 48 c7 c0 ff ff ff ff eb b8 0f 1f 00 f3 0f 1e fa 80 3d bd d5 0c 00 00 41 89 ca 74 10 b8 2c 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 71 c3 55 48 83 ec 30 44 89 4c 24 2c 4c 89 44 +[ 136.616381] RSP: 002b:00007ffde1796d68 EFLAGS: 00000202 ORIG_RAX: 000000000000002c +[ 136.616388] RAX: ffffffffffffffda RBX: 000055d7bd89f2a0 RCX: 00007fd30ba7b047 +[ 136.616392] RDX: 0000000000000028 RSI: 000055d7bd89f3b0 RDI: 0000000000000003 +[ 136.616396] RBP: 00007ffde1796e10 R08: 00007fd30bb4e200 R09: 000000000000000c +[ 136.616399] R10: 0000000000000000 R11: 0000000000000202 R12: 000055d7bd89f340 +[ 136.616403] R13: 000055d7bd89f3b0 R14: 000055d78943f200 R15: 0000000000000000 + +Fixes: 02cbfba1add5 ("idpf: add ethtool callbacks") +Reviewed-by: Ahmed Zaki +Signed-off-by: Michal Swiatkowski +Reviewed-by: Simon Horman +Tested-by: Samuel Salin +Signed-off-by: Tony Nguyen +Signed-off-by: Sasha Levin +--- + drivers/net/ethernet/intel/idpf/idpf_ethtool.c | 4 ++-- + 1 file changed, 2 insertions(+), 2 deletions(-) + +diff --git a/drivers/net/ethernet/intel/idpf/idpf_ethtool.c b/drivers/net/ethernet/intel/idpf/idpf_ethtool.c +index 59b1a1a099967..f72420cf68216 100644 +--- a/drivers/net/ethernet/intel/idpf/idpf_ethtool.c ++++ b/drivers/net/ethernet/intel/idpf/idpf_ethtool.c +@@ -46,7 +46,7 @@ static u32 idpf_get_rxfh_key_size(struct net_device *netdev) + struct idpf_vport_user_config_data *user_config; + + if (!idpf_is_cap_ena_all(np->adapter, IDPF_RSS_CAPS, IDPF_CAP_RSS)) +- return -EOPNOTSUPP; ++ return 0; + + user_config = &np->adapter->vport_config[np->vport_idx]->user_config; + +@@ -65,7 +65,7 @@ static u32 idpf_get_rxfh_indir_size(struct net_device *netdev) + struct idpf_vport_user_config_data *user_config; + + if (!idpf_is_cap_ena_all(np->adapter, IDPF_RSS_CAPS, IDPF_CAP_RSS)) +- return -EOPNOTSUPP; ++ return 0; + + user_config = &np->adapter->vport_config[np->vport_idx]->user_config; + +-- +2.39.5 + diff --git a/queue-6.15/igc-disable-l1.2-pci-e-link-substate-to-avoid-perfor.patch b/queue-6.15/igc-disable-l1.2-pci-e-link-substate-to-avoid-perfor.patch new file mode 100644 index 0000000000..4ffc63ac72 --- /dev/null +++ b/queue-6.15/igc-disable-l1.2-pci-e-link-substate-to-avoid-perfor.patch @@ -0,0 +1,66 @@ +From 89ec734a1da7c4cb45107b21daf2a5b8032979db Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Wed, 11 Jun 2025 15:52:54 +0300 +Subject: igc: disable L1.2 PCI-E link substate to avoid performance issue + +From: Vitaly Lifshits + +[ Upstream commit 0325143b59c6c6d79987afc57d2456e7a20d13b7 ] + +I226 devices advertise support for the PCI-E link L1.2 substate. However, +due to a hardware limitation, the exit latency from this low-power state +is longer than the packet buffer can tolerate under high traffic +conditions. This can lead to packet loss and degraded performance. + +To mitigate this, disable the L1.2 substate. The increased power draw +between L1.1 and L1.2 is insignificant. + +Fixes: 43546211738e ("igc: Add new device ID's") +Link: https://lore.kernel.org/intel-wired-lan/15248b4f-3271-42dd-8e35-02bfc92b25e1@intel.com +Signed-off-by: Vitaly Lifshits +Reviewed-by: Aleksandr Loktionov +Tested-by: Mor Bar-Gabay +Signed-off-by: Tony Nguyen +Signed-off-by: Sasha Levin +--- + drivers/net/ethernet/intel/igc/igc_main.c | 10 ++++++++++ + 1 file changed, 10 insertions(+) + +diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c +index b1669d7cf4359..7d8cc783b2282 100644 +--- a/drivers/net/ethernet/intel/igc/igc_main.c ++++ b/drivers/net/ethernet/intel/igc/igc_main.c +@@ -7046,6 +7046,10 @@ static int igc_probe(struct pci_dev *pdev, + adapter->port_num = hw->bus.func; + adapter->msg_enable = netif_msg_init(debug, DEFAULT_MSG_ENABLE); + ++ /* Disable ASPM L1.2 on I226 devices to avoid packet loss */ ++ if (igc_is_device_id_i226(hw)) ++ pci_disable_link_state(pdev, PCIE_LINK_STATE_L1_2); ++ + err = pci_save_state(pdev); + if (err) + goto err_ioremap; +@@ -7426,6 +7430,9 @@ static int __igc_resume(struct device *dev, bool rpm) + pci_enable_wake(pdev, PCI_D3hot, 0); + pci_enable_wake(pdev, PCI_D3cold, 0); + ++ if (igc_is_device_id_i226(hw)) ++ pci_disable_link_state(pdev, PCIE_LINK_STATE_L1_2); ++ + if (igc_init_interrupt_scheme(adapter, true)) { + netdev_err(netdev, "Unable to allocate memory for queues\n"); + return -ENOMEM; +@@ -7551,6 +7558,9 @@ static pci_ers_result_t igc_io_slot_reset(struct pci_dev *pdev) + pci_enable_wake(pdev, PCI_D3hot, 0); + pci_enable_wake(pdev, PCI_D3cold, 0); + ++ if (igc_is_device_id_i226(hw)) ++ pci_disable_link_state_locked(pdev, PCIE_LINK_STATE_L1_2); ++ + /* In case of PCI error, adapter loses its HW address + * so we should re-assign it here. + */ +-- +2.39.5 + diff --git a/queue-6.15/lib-test_objagg-set-error-message-in-check_expect_hi.patch b/queue-6.15/lib-test_objagg-set-error-message-in-check_expect_hi.patch new file mode 100644 index 0000000000..4f87f5b46f --- /dev/null +++ b/queue-6.15/lib-test_objagg-set-error-message-in-check_expect_hi.patch @@ -0,0 +1,50 @@ +From 9169070f8a73ef54f34f9cdbdf619a8236fbef2b Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Mon, 30 Jun 2025 14:36:40 -0500 +Subject: lib: test_objagg: Set error message in check_expect_hints_stats() + +From: Dan Carpenter + +[ Upstream commit e6ed134a4ef592fe1fd0cafac9683813b3c8f3e8 ] + +Smatch complains that the error message isn't set in the caller: + + lib/test_objagg.c:923 test_hints_case2() + error: uninitialized symbol 'errmsg'. + +This static checker warning only showed up after a recent refactoring +but the bug dates back to when the code was originally added. This +likely doesn't affect anything in real life. + +Reported-by: kernel test robot +Closes: https://lore.kernel.org/r/202506281403.DsuyHFTZ-lkp@intel.com/ +Fixes: 0a020d416d0a ("lib: introduce initial implementation of object aggregation manager") +Signed-off-by: Dan Carpenter +Reviewed-by: Ido Schimmel +Reviewed-by: Simon Horman +Link: https://patch.msgid.link/8548f423-2e3b-4bb7-b816-5041de2762aa@sabinyo.mountain +Signed-off-by: Jakub Kicinski +Signed-off-by: Sasha Levin +--- + lib/test_objagg.c | 4 +++- + 1 file changed, 3 insertions(+), 1 deletion(-) + +diff --git a/lib/test_objagg.c b/lib/test_objagg.c +index d34df4306b874..222b39fc2629e 100644 +--- a/lib/test_objagg.c ++++ b/lib/test_objagg.c +@@ -899,8 +899,10 @@ static int check_expect_hints_stats(struct objagg_hints *objagg_hints, + int err; + + stats = objagg_hints_stats_get(objagg_hints); +- if (IS_ERR(stats)) ++ if (IS_ERR(stats)) { ++ *errmsg = "objagg_hints_stats_get() failed."; + return PTR_ERR(stats); ++ } + err = __check_expect_stats(stats, expect_stats, errmsg); + objagg_stats_put(stats); + return err; +-- +2.39.5 + diff --git a/queue-6.15/module-provide-export_symbol_gpl_for_modules-helper.patch b/queue-6.15/module-provide-export_symbol_gpl_for_modules-helper.patch new file mode 100644 index 0000000000..3fde723475 --- /dev/null +++ b/queue-6.15/module-provide-export_symbol_gpl_for_modules-helper.patch @@ -0,0 +1,110 @@ +From 7edbf0d1181068b6369c3d886149b41bbb881c4c Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Fri, 2 May 2025 16:12:09 +0200 +Subject: module: Provide EXPORT_SYMBOL_GPL_FOR_MODULES() helper + +From: Peter Zijlstra + +[ Upstream commit 707f853d7fa3ce323a6875487890c213e34d81a0 ] + +Helper macro to more easily limit the export of a symbol to a given +list of modules. + +Eg: + + EXPORT_SYMBOL_GPL_FOR_MODULES(preempt_notifier_inc, "kvm"); + +will limit the use of said function to kvm.ko, any other module trying +to use this symbol will refure to load (and get modpost build +failures). + +Requested-by: Masahiro Yamada +Requested-by: Christoph Hellwig +Signed-off-by: Peter Zijlstra +Reviewed-by: Petr Pavlu +Signed-off-by: Masahiro Yamada +Stable-dep-of: cbe4134ea4bc ("fs: export anon_inode_make_secure_inode() and fix secretmem LSM bypass") +Signed-off-by: Sasha Levin +--- + Documentation/core-api/symbol-namespaces.rst | 22 ++++++++++++++++++++ + include/linux/export.h | 12 +++++++++-- + 2 files changed, 32 insertions(+), 2 deletions(-) + +diff --git a/Documentation/core-api/symbol-namespaces.rst b/Documentation/core-api/symbol-namespaces.rst +index 06f766a6aab24..c6f59c5e25648 100644 +--- a/Documentation/core-api/symbol-namespaces.rst ++++ b/Documentation/core-api/symbol-namespaces.rst +@@ -28,6 +28,9 @@ kernel. As of today, modules that make use of symbols exported into namespaces, + are required to import the namespace. Otherwise the kernel will, depending on + its configuration, reject loading the module or warn about a missing import. + ++Additionally, it is possible to put symbols into a module namespace, strictly ++limiting which modules are allowed to use these symbols. ++ + 2. How to define Symbol Namespaces + ================================== + +@@ -83,6 +86,22 @@ unit as preprocessor statement. The above example would then read:: + within the corresponding compilation unit before the #include for + . Typically it's placed before the first #include statement. + ++2.3 Using the EXPORT_SYMBOL_GPL_FOR_MODULES() macro ++=================================================== ++ ++Symbols exported using this macro are put into a module namespace. This ++namespace cannot be imported. ++ ++The macro takes a comma separated list of module names, allowing only those ++modules to access this symbol. Simple tail-globs are supported. ++ ++For example: ++ ++ EXPORT_SYMBOL_GPL_FOR_MODULES(preempt_notifier_inc, "kvm,kvm-*") ++ ++will limit usage of this symbol to modules whoes name matches the given ++patterns. ++ + 3. How to use Symbols exported in Namespaces + ============================================ + +@@ -154,3 +173,6 @@ in-tree modules:: + You can also run nsdeps for external module builds. A typical usage is:: + + $ make -C M=$PWD nsdeps ++ ++Note: it will happily generate an import statement for the module namespace; ++which will not work and generates build and runtime failures. +diff --git a/include/linux/export.h b/include/linux/export.h +index a8c23d945634b..f35d03b4113b1 100644 +--- a/include/linux/export.h ++++ b/include/linux/export.h +@@ -24,11 +24,17 @@ + .long sym + #endif + +-#define ___EXPORT_SYMBOL(sym, license, ns) \ ++/* ++ * LLVM integrated assembler cam merge adjacent string literals (like ++ * C and GNU-as) passed to '.ascii', but not to '.asciz' and chokes on: ++ * ++ * .asciz "MODULE_" "kvm" ; ++ */ ++#define ___EXPORT_SYMBOL(sym, license, ns...) \ + .section ".export_symbol","a" ASM_NL \ + __export_symbol_##sym: ASM_NL \ + .asciz license ASM_NL \ +- .asciz ns ASM_NL \ ++ .ascii ns "\0" ASM_NL \ + __EXPORT_SYMBOL_REF(sym) ASM_NL \ + .previous + +@@ -85,4 +91,6 @@ + #define EXPORT_SYMBOL_NS(sym, ns) __EXPORT_SYMBOL(sym, "", ns) + #define EXPORT_SYMBOL_NS_GPL(sym, ns) __EXPORT_SYMBOL(sym, "GPL", ns) + ++#define EXPORT_SYMBOL_GPL_FOR_MODULES(sym, mods) __EXPORT_SYMBOL(sym, "GPL", "module:" mods) ++ + #endif /* _LINUX_EXPORT_H */ +-- +2.39.5 + diff --git a/queue-6.15/mtd-spinand-fix-memory-leak-of-ecc-engine-conf.patch b/queue-6.15/mtd-spinand-fix-memory-leak-of-ecc-engine-conf.patch new file mode 100644 index 0000000000..d92f19d512 --- /dev/null +++ b/queue-6.15/mtd-spinand-fix-memory-leak-of-ecc-engine-conf.patch @@ -0,0 +1,59 @@ +From 10bb6b3d5d530c509870cabf37db373076bae563 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Wed, 18 Jun 2025 13:35:16 +0200 +Subject: mtd: spinand: fix memory leak of ECC engine conf + +From: Pablo Martin-Gomez + +[ Upstream commit 6463cbe08b0cbf9bba8763306764f5fd643023e1 ] + +Memory allocated for the ECC engine conf is not released during spinand +cleanup. Below kmemleak trace is seen for this memory leak: + +unreferenced object 0xffffff80064f00e0 (size 8): + comm "swapper/0", pid 1, jiffies 4294937458 + hex dump (first 8 bytes): + 00 00 00 00 00 00 00 00 ........ + backtrace (crc 0): + kmemleak_alloc+0x30/0x40 + __kmalloc_cache_noprof+0x208/0x3c0 + spinand_ondie_ecc_init_ctx+0x114/0x200 + nand_ecc_init_ctx+0x70/0xa8 + nanddev_ecc_engine_init+0xec/0x27c + spinand_probe+0xa2c/0x1620 + spi_mem_probe+0x130/0x21c + spi_probe+0xf0/0x170 + really_probe+0x17c/0x6e8 + __driver_probe_device+0x17c/0x21c + driver_probe_device+0x58/0x180 + __device_attach_driver+0x15c/0x1f8 + bus_for_each_drv+0xec/0x150 + __device_attach+0x188/0x24c + device_initial_probe+0x10/0x20 + bus_probe_device+0x11c/0x160 + +Fix the leak by calling nanddev_ecc_engine_cleanup() inside +spinand_cleanup(). + +Signed-off-by: Pablo Martin-Gomez +Signed-off-by: Miquel Raynal +Signed-off-by: Sasha Levin +--- + drivers/mtd/nand/spi/core.c | 1 + + 1 file changed, 1 insertion(+) + +diff --git a/drivers/mtd/nand/spi/core.c b/drivers/mtd/nand/spi/core.c +index d16e42cf8faeb..303aede6e5aa4 100644 +--- a/drivers/mtd/nand/spi/core.c ++++ b/drivers/mtd/nand/spi/core.c +@@ -1585,6 +1585,7 @@ static void spinand_cleanup(struct spinand_device *spinand) + { + struct nand_device *nand = spinand_to_nand(spinand); + ++ nanddev_ecc_engine_cleanup(nand); + nanddev_cleanup(nand); + spinand_manufacturer_cleanup(spinand); + kfree(spinand->databuf); +-- +2.39.5 + diff --git a/queue-6.15/net-ipv4-fix-stat-increase-when-udp-early-demux-drop.patch b/queue-6.15/net-ipv4-fix-stat-increase-when-udp-early-demux-drop.patch new file mode 100644 index 0000000000..58c1d7d050 --- /dev/null +++ b/queue-6.15/net-ipv4-fix-stat-increase-when-udp-early-demux-drop.patch @@ -0,0 +1,60 @@ +From bc98ce92823efa9a6e7b450a34a46fa274b0b505 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Tue, 1 Jul 2025 09:49:34 +0200 +Subject: net: ipv4: fix stat increase when udp early demux drops the packet + +From: Antoine Tenart + +[ Upstream commit c2a2ff6b4db55647575260bf2227b0e09d46addb ] + +udp_v4_early_demux now returns drop reasons as it either returns 0 or +ip_mc_validate_source, which returns itself a drop reason. However its +use was not converted in ip_rcv_finish_core and the drop reason is +ignored, leading to potentially skipping increasing LINUX_MIB_IPRPFILTER +if the drop reason is SKB_DROP_REASON_IP_RPFILTER. + +This is a fix and we're not converting udp_v4_early_demux to explicitly +return a drop reason to ease backports; this can be done as a follow-up. + +Fixes: d46f827016d8 ("net: ip: make ip_mc_validate_source() return drop reason") +Cc: Menglong Dong +Reported-by: Sabrina Dubroca +Signed-off-by: Antoine Tenart +Reviewed-by: Sabrina Dubroca +Link: https://patch.msgid.link/20250701074935.144134-1-atenart@kernel.org +Signed-off-by: Jakub Kicinski +Signed-off-by: Sasha Levin +--- + net/ipv4/ip_input.c | 7 ++++--- + 1 file changed, 4 insertions(+), 3 deletions(-) + +diff --git a/net/ipv4/ip_input.c b/net/ipv4/ip_input.c +index 30a5e9460d006..5a49eb99e5c48 100644 +--- a/net/ipv4/ip_input.c ++++ b/net/ipv4/ip_input.c +@@ -319,8 +319,8 @@ static int ip_rcv_finish_core(struct net *net, + const struct sk_buff *hint) + { + const struct iphdr *iph = ip_hdr(skb); +- int err, drop_reason; + struct rtable *rt; ++ int drop_reason; + + if (ip_can_use_hint(skb, iph, hint)) { + drop_reason = ip_route_use_hint(skb, iph->daddr, iph->saddr, +@@ -345,9 +345,10 @@ static int ip_rcv_finish_core(struct net *net, + break; + case IPPROTO_UDP: + if (READ_ONCE(net->ipv4.sysctl_udp_early_demux)) { +- err = udp_v4_early_demux(skb); +- if (unlikely(err)) ++ drop_reason = udp_v4_early_demux(skb); ++ if (unlikely(drop_reason)) + goto drop_error; ++ drop_reason = SKB_DROP_REASON_NOT_SPECIFIED; + + /* must reload iph, skb->head might have changed */ + iph = ip_hdr(skb); +-- +2.39.5 + diff --git a/queue-6.15/net-sched-always-pass-notifications-when-child-class.patch b/queue-6.15/net-sched-always-pass-notifications-when-child-class.patch new file mode 100644 index 0000000000..997a399648 --- /dev/null +++ b/queue-6.15/net-sched-always-pass-notifications-when-child-class.patch @@ -0,0 +1,109 @@ +From fbdf56739a622d95de1c828057ecd7335f41499e Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Mon, 30 Jun 2025 15:27:30 +0200 +Subject: net/sched: Always pass notifications when child class becomes empty + +From: Lion Ackermann + +[ Upstream commit 103406b38c600fec1fe375a77b27d87e314aea09 ] + +Certain classful qdiscs may invoke their classes' dequeue handler on an +enqueue operation. This may unexpectedly empty the child qdisc and thus +make an in-flight class passive via qlen_notify(). Most qdiscs do not +expect such behaviour at this point in time and may re-activate the +class eventually anyways which will lead to a use-after-free. + +The referenced fix commit attempted to fix this behavior for the HFSC +case by moving the backlog accounting around, though this turned out to +be incomplete since the parent's parent may run into the issue too. +The following reproducer demonstrates this use-after-free: + + tc qdisc add dev lo root handle 1: drr + tc filter add dev lo parent 1: basic classid 1:1 + tc class add dev lo parent 1: classid 1:1 drr + tc qdisc add dev lo parent 1:1 handle 2: hfsc def 1 + tc class add dev lo parent 2: classid 2:1 hfsc rt m1 8 d 1 m2 0 + tc qdisc add dev lo parent 2:1 handle 3: netem + tc qdisc add dev lo parent 3:1 handle 4: blackhole + + echo 1 | socat -u STDIN UDP4-DATAGRAM:127.0.0.1:8888 + tc class delete dev lo classid 1:1 + echo 1 | socat -u STDIN UDP4-DATAGRAM:127.0.0.1:8888 + +Since backlog accounting issues leading to a use-after-frees on stale +class pointers is a recurring pattern at this point, this patch takes +a different approach. Instead of trying to fix the accounting, the patch +ensures that qdisc_tree_reduce_backlog always calls qlen_notify when +the child qdisc is empty. This solves the problem because deletion of +qdiscs always involves a call to qdisc_reset() and / or +qdisc_purge_queue() which ultimately resets its qlen to 0 thus causing +the following qdisc_tree_reduce_backlog() to report to the parent. Note +that this may call qlen_notify on passive classes multiple times. This +is not a problem after the recent patch series that made all the +classful qdiscs qlen_notify() handlers idempotent. + +Fixes: 3f981138109f ("sch_hfsc: Fix qlen accounting bug when using peek in hfsc_enqueue()") +Signed-off-by: Lion Ackermann +Reviewed-by: Jamal Hadi Salim +Acked-by: Cong Wang +Acked-by: Jamal Hadi Salim +Link: https://patch.msgid.link/d912cbd7-193b-4269-9857-525bee8bbb6a@gmail.com +Signed-off-by: Jakub Kicinski +Signed-off-by: Sasha Levin +--- + net/sched/sch_api.c | 19 +++++-------------- + 1 file changed, 5 insertions(+), 14 deletions(-) + +diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c +index f74a097f54ae7..d58921ffcf35e 100644 +--- a/net/sched/sch_api.c ++++ b/net/sched/sch_api.c +@@ -779,15 +779,12 @@ static u32 qdisc_alloc_handle(struct net_device *dev) + + void qdisc_tree_reduce_backlog(struct Qdisc *sch, int n, int len) + { +- bool qdisc_is_offloaded = sch->flags & TCQ_F_OFFLOADED; + const struct Qdisc_class_ops *cops; + unsigned long cl; + u32 parentid; + bool notify; + int drops; + +- if (n == 0 && len == 0) +- return; + drops = max_t(int, n, 0); + rcu_read_lock(); + while ((parentid = sch->parent)) { +@@ -796,17 +793,8 @@ void qdisc_tree_reduce_backlog(struct Qdisc *sch, int n, int len) + + if (sch->flags & TCQ_F_NOPARENT) + break; +- /* Notify parent qdisc only if child qdisc becomes empty. +- * +- * If child was empty even before update then backlog +- * counter is screwed and we skip notification because +- * parent class is already passive. +- * +- * If the original child was offloaded then it is allowed +- * to be seem as empty, so the parent is notified anyway. +- */ +- notify = !sch->q.qlen && !WARN_ON_ONCE(!n && +- !qdisc_is_offloaded); ++ /* Notify parent qdisc only if child qdisc becomes empty. */ ++ notify = !sch->q.qlen; + /* TODO: perform the search on a per txq basis */ + sch = qdisc_lookup_rcu(qdisc_dev(sch), TC_H_MAJ(parentid)); + if (sch == NULL) { +@@ -815,6 +803,9 @@ void qdisc_tree_reduce_backlog(struct Qdisc *sch, int n, int len) + } + cops = sch->ops->cl_ops; + if (notify && cops->qlen_notify) { ++ /* Note that qlen_notify must be idempotent as it may get called ++ * multiple times. ++ */ + cl = cops->find(sch, parentid); + cops->qlen_notify(sch, cl); + } +-- +2.39.5 + diff --git a/queue-6.15/net-usb-lan78xx-fix-warn-in-__netif_napi_del_locked-.patch b/queue-6.15/net-usb-lan78xx-fix-warn-in-__netif_napi_del_locked-.patch new file mode 100644 index 0000000000..5c0293fd27 --- /dev/null +++ b/queue-6.15/net-usb-lan78xx-fix-warn-in-__netif_napi_del_locked-.patch @@ -0,0 +1,96 @@ +From f224950f48ae2fa921f941b2c9ffc1b146057f77 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Fri, 27 Jun 2025 07:13:46 +0200 +Subject: net: usb: lan78xx: fix WARN in __netif_napi_del_locked on disconnect + +From: Oleksij Rempel + +[ Upstream commit 6c7ffc9af7186ed79403a3ffee9a1e5199fc7450 ] + +Remove redundant netif_napi_del() call from disconnect path. + +A WARN may be triggered in __netif_napi_del_locked() during USB device +disconnect: + + WARNING: CPU: 0 PID: 11 at net/core/dev.c:7417 __netif_napi_del_locked+0x2b4/0x350 + +This happens because netif_napi_del() is called in the disconnect path while +NAPI is still enabled. However, it is not necessary to call netif_napi_del() +explicitly, since unregister_netdev() will handle NAPI teardown automatically +and safely. Removing the redundant call avoids triggering the warning. + +Full trace: + lan78xx 1-1:1.0 enu1: Failed to read register index 0x000000c4. ret = -ENODEV + lan78xx 1-1:1.0 enu1: Failed to set MAC down with error -ENODEV + lan78xx 1-1:1.0 enu1: Link is Down + lan78xx 1-1:1.0 enu1: Failed to read register index 0x00000120. ret = -ENODEV + ------------[ cut here ]------------ + WARNING: CPU: 0 PID: 11 at net/core/dev.c:7417 __netif_napi_del_locked+0x2b4/0x350 + Modules linked in: flexcan can_dev fuse + CPU: 0 UID: 0 PID: 11 Comm: kworker/0:1 Not tainted 6.16.0-rc2-00624-ge926949dab03 #9 PREEMPT + Hardware name: SKOV IMX8MP CPU revC - bd500 (DT) + Workqueue: usb_hub_wq hub_event + pstate: 60000005 (nZCv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--) + pc : __netif_napi_del_locked+0x2b4/0x350 + lr : __netif_napi_del_locked+0x7c/0x350 + sp : ffffffc085b673c0 + x29: ffffffc085b673c0 x28: ffffff800b7f2000 x27: ffffff800b7f20d8 + x26: ffffff80110bcf58 x25: ffffff80110bd978 x24: 1ffffff0022179eb + x23: ffffff80110bc000 x22: ffffff800b7f5000 x21: ffffff80110bc000 + x20: ffffff80110bcf38 x19: ffffff80110bcf28 x18: dfffffc000000000 + x17: ffffffc081578940 x16: ffffffc08284cee0 x15: 0000000000000028 + x14: 0000000000000006 x13: 0000000000040000 x12: ffffffb0022179e8 + x11: 1ffffff0022179e7 x10: ffffffb0022179e7 x9 : dfffffc000000000 + x8 : 0000004ffdde8619 x7 : ffffff80110bcf3f x6 : 0000000000000001 + x5 : ffffff80110bcf38 x4 : ffffff80110bcf38 x3 : 0000000000000000 + x2 : 0000000000000000 x1 : 1ffffff0022179e7 x0 : 0000000000000000 + Call trace: + __netif_napi_del_locked+0x2b4/0x350 (P) + lan78xx_disconnect+0xf4/0x360 + usb_unbind_interface+0x158/0x718 + device_remove+0x100/0x150 + device_release_driver_internal+0x308/0x478 + device_release_driver+0x1c/0x30 + bus_remove_device+0x1a8/0x368 + device_del+0x2e0/0x7b0 + usb_disable_device+0x244/0x540 + usb_disconnect+0x220/0x758 + hub_event+0x105c/0x35e0 + process_one_work+0x760/0x17b0 + worker_thread+0x768/0xce8 + kthread+0x3bc/0x690 + ret_from_fork+0x10/0x20 + irq event stamp: 211604 + hardirqs last enabled at (211603): [] _raw_spin_unlock_irqrestore+0x84/0x98 + hardirqs last disabled at (211604): [] el1_dbg+0x24/0x80 + softirqs last enabled at (211296): [] handle_softirqs+0x820/0xbc8 + softirqs last disabled at (210993): [] __do_softirq+0x18/0x20 + ---[ end trace 0000000000000000 ]--- + lan78xx 1-1:1.0 enu1: failed to kill vid 0081/0 + +Fixes: ec4c7e12396b ("lan78xx: Introduce NAPI polling support") +Suggested-by: Jakub Kicinski +Signed-off-by: Oleksij Rempel +Link: https://patch.msgid.link/20250627051346.276029-1-o.rempel@pengutronix.de +Signed-off-by: Jakub Kicinski +Signed-off-by: Sasha Levin +--- + drivers/net/usb/lan78xx.c | 2 -- + 1 file changed, 2 deletions(-) + +diff --git a/drivers/net/usb/lan78xx.c b/drivers/net/usb/lan78xx.c +index e4f1663b62047..5f014b8626853 100644 +--- a/drivers/net/usb/lan78xx.c ++++ b/drivers/net/usb/lan78xx.c +@@ -4344,8 +4344,6 @@ static void lan78xx_disconnect(struct usb_interface *intf) + if (!dev) + return; + +- netif_napi_del(&dev->napi); +- + udev = interface_to_usbdev(intf); + net = dev->net; + +-- +2.39.5 + diff --git a/queue-6.15/netfs-fix-double-put-of-request.patch b/queue-6.15/netfs-fix-double-put-of-request.patch new file mode 100644 index 0000000000..c448cbfb0c --- /dev/null +++ b/queue-6.15/netfs-fix-double-put-of-request.patch @@ -0,0 +1,86 @@ +From a4ed508c6a2a97ab6318eab7953d224a5929be22 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Tue, 1 Jul 2025 17:38:37 +0100 +Subject: netfs: Fix double put of request + +From: David Howells + +[ Upstream commit 9df7b5ebead649b00bf9a53a798e4bf83a1318fd ] + +If a netfs request finishes during the pause loop, it will have the ref +that belongs to the IN_PROGRESS flag removed at that point - however, if it +then goes to the final wait loop, that will *also* put the ref because it +sees that the IN_PROGRESS flag is clear and incorrectly assumes that this +happened when it called the collector. + +In fact, since IN_PROGRESS is clear, we shouldn't call the collector again +since it's done all the cleanup, such as calling ->ki_complete(). + +Fix this by making netfs_collect_in_app() just return, indicating that +we're done if IN_PROGRESS is removed. + +Fixes: 2b1424cd131c ("netfs: Fix wait/wake to be consistent about the waitqueue used") +Signed-off-by: David Howells +Link: https://lore.kernel.org/20250701163852.2171681-3-dhowells@redhat.com +Tested-by: Steve French +Reviewed-by: Paulo Alcantara +cc: Steve French +cc: netfs@lists.linux.dev +cc: linux-fsdevel@vger.kernel.org +cc: linux-cifs@vger.kernel.org +Signed-off-by: Christian Brauner +Signed-off-by: Sasha Levin +--- + fs/netfs/misc.c | 9 +++++++-- + include/trace/events/netfs.h | 1 + + 2 files changed, 8 insertions(+), 2 deletions(-) + +diff --git a/fs/netfs/misc.c b/fs/netfs/misc.c +index d8b1a279dbda9..8b1c11ef32aa5 100644 +--- a/fs/netfs/misc.c ++++ b/fs/netfs/misc.c +@@ -383,6 +383,11 @@ static int netfs_collect_in_app(struct netfs_io_request *rreq, + { + bool need_collect = false, inactive = true, done = true; + ++ if (!test_bit(NETFS_RREQ_IN_PROGRESS, &rreq->flags)) { ++ trace_netfs_rreq(rreq, netfs_rreq_trace_recollect); ++ return 1; /* Done */ ++ } ++ + for (int i = 0; i < NR_IO_STREAMS; i++) { + struct netfs_io_subrequest *subreq; + struct netfs_io_stream *stream = &rreq->io_streams[i]; +@@ -442,7 +447,7 @@ static ssize_t netfs_wait_for_in_progress(struct netfs_io_request *rreq, + case 1: + goto all_collected; + case 2: +- if (!netfs_check_rreq_in_progress(rreq)) ++ if (!test_bit(NETFS_RREQ_IN_PROGRESS, &rreq->flags)) + break; + cond_resched(); + continue; +@@ -512,7 +517,7 @@ static void netfs_wait_for_pause(struct netfs_io_request *rreq, + case 1: + goto all_collected; + case 2: +- if (!netfs_check_rreq_in_progress(rreq) || ++ if (!test_bit(NETFS_RREQ_IN_PROGRESS, &rreq->flags) || + !test_bit(NETFS_RREQ_PAUSE, &rreq->flags)) + break; + cond_resched(); +diff --git a/include/trace/events/netfs.h b/include/trace/events/netfs.h +index 4175eec40048a..ecc1b852661e3 100644 +--- a/include/trace/events/netfs.h ++++ b/include/trace/events/netfs.h +@@ -56,6 +56,7 @@ + EM(netfs_rreq_trace_dirty, "DIRTY ") \ + EM(netfs_rreq_trace_done, "DONE ") \ + EM(netfs_rreq_trace_free, "FREE ") \ ++ EM(netfs_rreq_trace_recollect, "RECLLCT") \ + EM(netfs_rreq_trace_redirty, "REDIRTY") \ + EM(netfs_rreq_trace_resubmit, "RESUBMT") \ + EM(netfs_rreq_trace_set_abandon, "S-ABNDN") \ +-- +2.39.5 + diff --git a/queue-6.15/netfs-fix-hang-due-to-missing-case-in-final-dio-read.patch b/queue-6.15/netfs-fix-hang-due-to-missing-case-in-final-dio-read.patch new file mode 100644 index 0000000000..2ec1423fda --- /dev/null +++ b/queue-6.15/netfs-fix-hang-due-to-missing-case-in-final-dio-read.patch @@ -0,0 +1,79 @@ +From d1f14be026b0f656f25aa96173bf700a952be41d Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Tue, 1 Jul 2025 17:38:36 +0100 +Subject: netfs: Fix hang due to missing case in final DIO read result + collection + +From: David Howells + +[ Upstream commit da8cf4bd458722d090a788c6e581eeb72695c62f ] + +When doing a DIO read, if the subrequests we issue fail and cause the +request PAUSE flag to be set to put a pause on subrequest generation, we +may complete collection of the subrequests (possibly discarding them) prior +to the ALL_QUEUED flags being set. + +In such a case, netfs_read_collection() doesn't see ALL_QUEUED being set +after netfs_collect_read_results() returns and will just return to the app +(the collector can be seen unpausing the generator in the trace log). + +The subrequest generator can then set ALL_QUEUED and the app thread reaches +netfs_wait_for_request(). This causes netfs_collect_in_app() to be called +to see if we're done yet, but there's missing case here. + +netfs_collect_in_app() will see that a thread is active and set inactive to +false, but won't see any subrequests in the read stream, and so won't set +need_collect to true. The function will then just return 0, indicating +that the caller should just sleep until further activity (which won't be +forthcoming) occurs. + +Fix this by making netfs_collect_in_app() check to see if an active thread +is complete - i.e. that ALL_QUEUED is set and the subrequests list is empty +- and to skip the sleep return path. The collector will then be called +which will clear the request IN_PROGRESS flag, allowing the app to +progress. + +Fixes: 2b1424cd131c ("netfs: Fix wait/wake to be consistent about the waitqueue used") +Reported-by: Steve French +Signed-off-by: David Howells +Link: https://lore.kernel.org/20250701163852.2171681-2-dhowells@redhat.com +Tested-by: Steve French +Reviewed-by: Paulo Alcantara +cc: linux-cifs@vger.kernel.org +cc: netfs@lists.linux.dev +cc: linux-fsdevel@vger.kernel.org +Signed-off-by: Christian Brauner +Signed-off-by: Sasha Levin +--- + fs/netfs/misc.c | 6 ++++-- + 1 file changed, 4 insertions(+), 2 deletions(-) + +diff --git a/fs/netfs/misc.c b/fs/netfs/misc.c +index 43b67a28a8fa0..0a54b12034868 100644 +--- a/fs/netfs/misc.c ++++ b/fs/netfs/misc.c +@@ -381,7 +381,7 @@ void netfs_wait_for_in_progress_stream(struct netfs_io_request *rreq, + static int netfs_collect_in_app(struct netfs_io_request *rreq, + bool (*collector)(struct netfs_io_request *rreq)) + { +- bool need_collect = false, inactive = true; ++ bool need_collect = false, inactive = true, done = true; + + for (int i = 0; i < NR_IO_STREAMS; i++) { + struct netfs_io_subrequest *subreq; +@@ -400,9 +400,11 @@ static int netfs_collect_in_app(struct netfs_io_request *rreq, + need_collect = true; + break; + } ++ if (subreq || !test_bit(NETFS_RREQ_ALL_QUEUED, &rreq->flags)) ++ done = false; + } + +- if (!need_collect && !inactive) ++ if (!need_collect && !inactive && !done) + return 0; /* Sleep */ + + __set_current_state(TASK_RUNNING); +-- +2.39.5 + diff --git a/queue-6.15/netfs-fix-i_size-updating.patch b/queue-6.15/netfs-fix-i_size-updating.patch new file mode 100644 index 0000000000..94fd48b2da --- /dev/null +++ b/queue-6.15/netfs-fix-i_size-updating.patch @@ -0,0 +1,86 @@ +From 93c34854601d7a7ada0c3993dd9c41783532b919 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Tue, 1 Jul 2025 17:38:45 +0100 +Subject: netfs: Fix i_size updating + +From: David Howells + +[ Upstream commit 2e0658940d90a3dc130bb3b7f75bae9f4100e01f ] + +Fix the updating of i_size, particularly in regard to the completion of DIO +writes and especially async DIO writes by using a lock. + +The bug is triggered occasionally by the generic/207 xfstest as it chucks a +bunch of AIO DIO writes at the filesystem and then checks that fstat() +returns a reasonable st_size as each completes. + +The problem is that netfs is trying to do "if new_size > inode->i_size, +update inode->i_size" sort of thing but without a lock around it. + +This can be seen with cifs, but shouldn't be seen with kafs because kafs +serialises modification ops on the client whereas cifs sends the requests +to the server as they're generated and lets the server order them. + +Fixes: 153a9961b551 ("netfs: Implement unbuffered/DIO write support") +Signed-off-by: David Howells +Link: https://lore.kernel.org/20250701163852.2171681-11-dhowells@redhat.com +Reviewed-by: Paulo Alcantara (Red Hat) +cc: Steve French +cc: Paulo Alcantara +cc: linux-cifs@vger.kernel.org +cc: netfs@lists.linux.dev +cc: linux-fsdevel@vger.kernel.org +Signed-off-by: Christian Brauner +Signed-off-by: Sasha Levin +--- + fs/netfs/buffered_write.c | 2 ++ + fs/netfs/direct_write.c | 8 ++++++-- + 2 files changed, 8 insertions(+), 2 deletions(-) + +diff --git a/fs/netfs/buffered_write.c b/fs/netfs/buffered_write.c +index dbb544e183d13..9f22ff890a8cd 100644 +--- a/fs/netfs/buffered_write.c ++++ b/fs/netfs/buffered_write.c +@@ -64,6 +64,7 @@ static void netfs_update_i_size(struct netfs_inode *ctx, struct inode *inode, + return; + } + ++ spin_lock(&inode->i_lock); + i_size_write(inode, pos); + #if IS_ENABLED(CONFIG_FSCACHE) + fscache_update_cookie(ctx->cache, NULL, &pos); +@@ -77,6 +78,7 @@ static void netfs_update_i_size(struct netfs_inode *ctx, struct inode *inode, + DIV_ROUND_UP(pos, SECTOR_SIZE), + inode->i_blocks + add); + } ++ spin_unlock(&inode->i_lock); + } + + /** +diff --git a/fs/netfs/direct_write.c b/fs/netfs/direct_write.c +index fa9a5bf3c6d51..3efa5894b2c07 100644 +--- a/fs/netfs/direct_write.c ++++ b/fs/netfs/direct_write.c +@@ -14,13 +14,17 @@ static void netfs_cleanup_dio_write(struct netfs_io_request *wreq) + struct inode *inode = wreq->inode; + unsigned long long end = wreq->start + wreq->transferred; + +- if (!wreq->error && +- i_size_read(inode) < end) { ++ if (wreq->error || end <= i_size_read(inode)) ++ return; ++ ++ spin_lock(&inode->i_lock); ++ if (end > i_size_read(inode)) { + if (wreq->netfs_ops->update_i_size) + wreq->netfs_ops->update_i_size(inode, end); + else + i_size_write(inode, end); + } ++ spin_unlock(&inode->i_lock); + } + + /* +-- +2.39.5 + diff --git a/queue-6.15/netfs-fix-looping-in-wait-functions.patch b/queue-6.15/netfs-fix-looping-in-wait-functions.patch new file mode 100644 index 0000000000..395722bb51 --- /dev/null +++ b/queue-6.15/netfs-fix-looping-in-wait-functions.patch @@ -0,0 +1,88 @@ +From 0e7d1bed5de370dd8b9426c7097272d513d97c3b Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Tue, 1 Jul 2025 17:38:39 +0100 +Subject: netfs: Fix looping in wait functions + +From: David Howells + +[ Upstream commit 09623e3a14c1c5465124350cd227457c2b0fb017 ] + +netfs_wait_for_request() and netfs_wait_for_pause() can loop forever if +netfs_collect_in_app() returns 2, indicating that it wants to repeat +because the ALL_QUEUED flag isn't yet set and there are no subreqs left +that haven't been collected. + +The problem is that, unless collection is offloaded (OFFLOAD_COLLECTION), +we have to return to the application thread to continue and eventually set +ALL_QUEUED after pausing to deal with a retry - but we never get there. + +Fix this by inserting checks for the IN_PROGRESS and PAUSE flags as +appropriate before cycling round - and add cond_resched() for good measure. + +Fixes: 2b1424cd131c ("netfs: Fix wait/wake to be consistent about the waitqueue used") +Signed-off-by: David Howells +Link: https://lore.kernel.org/20250701163852.2171681-5-dhowells@redhat.com +Tested-by: Steve French +Reviewed-by: Paulo Alcantara +cc: netfs@lists.linux.dev +cc: linux-fsdevel@vger.kernel.org +Signed-off-by: Christian Brauner +Signed-off-by: Sasha Levin +--- + fs/netfs/misc.c | 15 +++++++++++---- + 1 file changed, 11 insertions(+), 4 deletions(-) + +diff --git a/fs/netfs/misc.c b/fs/netfs/misc.c +index 0a54b12034868..d8b1a279dbda9 100644 +--- a/fs/netfs/misc.c ++++ b/fs/netfs/misc.c +@@ -425,8 +425,8 @@ static int netfs_collect_in_app(struct netfs_io_request *rreq, + /* + * Wait for a request to complete, successfully or otherwise. + */ +-static ssize_t netfs_wait_for_request(struct netfs_io_request *rreq, +- bool (*collector)(struct netfs_io_request *rreq)) ++static ssize_t netfs_wait_for_in_progress(struct netfs_io_request *rreq, ++ bool (*collector)(struct netfs_io_request *rreq)) + { + DEFINE_WAIT(myself); + ssize_t ret; +@@ -442,6 +442,9 @@ static ssize_t netfs_wait_for_request(struct netfs_io_request *rreq, + case 1: + goto all_collected; + case 2: ++ if (!netfs_check_rreq_in_progress(rreq)) ++ break; ++ cond_resched(); + continue; + } + } +@@ -480,12 +483,12 @@ static ssize_t netfs_wait_for_request(struct netfs_io_request *rreq, + + ssize_t netfs_wait_for_read(struct netfs_io_request *rreq) + { +- return netfs_wait_for_request(rreq, netfs_read_collection); ++ return netfs_wait_for_in_progress(rreq, netfs_read_collection); + } + + ssize_t netfs_wait_for_write(struct netfs_io_request *rreq) + { +- return netfs_wait_for_request(rreq, netfs_write_collection); ++ return netfs_wait_for_in_progress(rreq, netfs_write_collection); + } + + /* +@@ -509,6 +512,10 @@ static void netfs_wait_for_pause(struct netfs_io_request *rreq, + case 1: + goto all_collected; + case 2: ++ if (!netfs_check_rreq_in_progress(rreq) || ++ !test_bit(NETFS_RREQ_PAUSE, &rreq->flags)) ++ break; ++ cond_resched(); + continue; + } + } +-- +2.39.5 + diff --git a/queue-6.15/netfs-fix-ref-leak-on-inserted-extra-subreq-in-write.patch b/queue-6.15/netfs-fix-ref-leak-on-inserted-extra-subreq-in-write.patch new file mode 100644 index 0000000000..82c10020f8 --- /dev/null +++ b/queue-6.15/netfs-fix-ref-leak-on-inserted-extra-subreq-in-write.patch @@ -0,0 +1,49 @@ +From 485ff606f148b501dd8add82035d705fcfb85daf Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Tue, 1 Jul 2025 17:38:40 +0100 +Subject: netfs: Fix ref leak on inserted extra subreq in write retry + +From: David Howells + +[ Upstream commit 97d8e8e52cb8ab3d7675880a92626d9a4332f7a6 ] + +The write-retry algorithm will insert extra subrequests into the list if it +can't get sufficient capacity to split the range that needs to be retried +into the sequence of subrequests it currently has (for instance, if the +cifs credit pool has fewer credits available than it did when the range was +originally divided). + +However, the allocator furnishes each new subreq with 2 refs and then +another is added for resubmission, causing one to be leaked. + +Fix this by replacing the ref-getting line with a neutral trace line. + +Fixes: 288ace2f57c9 ("netfs: New writeback implementation") +Signed-off-by: David Howells +Link: https://lore.kernel.org/20250701163852.2171681-6-dhowells@redhat.com +Tested-by: Steve French +Reviewed-by: Paulo Alcantara +cc: netfs@lists.linux.dev +cc: linux-fsdevel@vger.kernel.org +Signed-off-by: Christian Brauner +Signed-off-by: Sasha Levin +--- + fs/netfs/write_retry.c | 2 +- + 1 file changed, 1 insertion(+), 1 deletion(-) + +diff --git a/fs/netfs/write_retry.c b/fs/netfs/write_retry.c +index 9d1d8a8bab726..7158657061e98 100644 +--- a/fs/netfs/write_retry.c ++++ b/fs/netfs/write_retry.c +@@ -153,7 +153,7 @@ static void netfs_retry_write_stream(struct netfs_io_request *wreq, + trace_netfs_sreq_ref(wreq->debug_id, subreq->debug_index, + refcount_read(&subreq->ref), + netfs_sreq_trace_new); +- netfs_get_subrequest(subreq, netfs_sreq_trace_get_resubmit); ++ trace_netfs_sreq(subreq, netfs_sreq_trace_split); + + list_add(&subreq->rreq_link, &to->rreq_link); + to = list_next_entry(to, rreq_link); +-- +2.39.5 + diff --git a/queue-6.15/nfs-clean-up-proc-net-rpc-nfs-when-nfs_fs_proc_net_i.patch b/queue-6.15/nfs-clean-up-proc-net-rpc-nfs-when-nfs_fs_proc_net_i.patch new file mode 100644 index 0000000000..86a2c6fa2a --- /dev/null +++ b/queue-6.15/nfs-clean-up-proc-net-rpc-nfs-when-nfs_fs_proc_net_i.patch @@ -0,0 +1,139 @@ +From 1860fa5ac2f4ade0aa83177f0856f1b72aae4eef Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Thu, 12 Jun 2025 14:52:50 -0700 +Subject: nfs: Clean up /proc/net/rpc/nfs when nfs_fs_proc_net_init() fails. + +From: Kuniyuki Iwashima + +[ Upstream commit e8d6f3ab59468e230f3253efe5cb63efa35289f7 ] + +syzbot reported a warning below [1] following a fault injection in +nfs_fs_proc_net_init(). [0] + +When nfs_fs_proc_net_init() fails, /proc/net/rpc/nfs is not removed. + +Later, rpc_proc_exit() tries to remove /proc/net/rpc, and the warning +is logged as the directory is not empty. + +Let's handle the error of nfs_fs_proc_net_init() properly. + +[0]: +FAULT_INJECTION: forcing a failure. +name failslab, interval 1, probability 0, space 0, times 0 +CPU: 1 UID: 0 PID: 6120 Comm: syz.2.27 Not tainted 6.16.0-rc1-syzkaller-00010-g2c4a1f3fe03e #0 PREEMPT(full) +Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 05/07/2025 +Call Trace: + + dump_stack_lvl (lib/dump_stack.c:123) + should_fail_ex (lib/fault-inject.c:73 lib/fault-inject.c:174) + should_failslab (mm/failslab.c:46) + kmem_cache_alloc_noprof (mm/slub.c:4178 mm/slub.c:4204) + __proc_create (fs/proc/generic.c:427) + proc_create_reg (fs/proc/generic.c:554) + proc_create_net_data (fs/proc/proc_net.c:120) + nfs_fs_proc_net_init (fs/nfs/client.c:1409) + nfs_net_init (fs/nfs/inode.c:2600) + ops_init (net/core/net_namespace.c:138) + setup_net (net/core/net_namespace.c:443) + copy_net_ns (net/core/net_namespace.c:576) + create_new_namespaces (kernel/nsproxy.c:110) + unshare_nsproxy_namespaces (kernel/nsproxy.c:218 (discriminator 4)) + ksys_unshare (kernel/fork.c:3123) + __x64_sys_unshare (kernel/fork.c:3190) + do_syscall_64 (arch/x86/entry/syscall_64.c:63 arch/x86/entry/syscall_64.c:94) + entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:130) + + +[1]: +remove_proc_entry: removing non-empty directory 'net/rpc', leaking at least 'nfs' + WARNING: CPU: 1 PID: 6120 at fs/proc/generic.c:727 remove_proc_entry+0x45e/0x530 fs/proc/generic.c:727 +Modules linked in: +CPU: 1 UID: 0 PID: 6120 Comm: syz.2.27 Not tainted 6.16.0-rc1-syzkaller-00010-g2c4a1f3fe03e #0 PREEMPT(full) +Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 05/07/2025 + RIP: 0010:remove_proc_entry+0x45e/0x530 fs/proc/generic.c:727 +Code: 3c 02 00 0f 85 85 00 00 00 48 8b 93 d8 00 00 00 4d 89 f0 4c 89 e9 48 c7 c6 40 ba a2 8b 48 c7 c7 60 b9 a2 8b e8 33 81 1d ff 90 <0f> 0b 90 90 e9 5f fe ff ff e8 04 69 5e ff 90 48 b8 00 00 00 00 00 +RSP: 0018:ffffc90003637b08 EFLAGS: 00010282 +RAX: 0000000000000000 RBX: ffff88805f534140 RCX: ffffffff817a92c8 +RDX: ffff88807da99e00 RSI: ffffffff817a92d5 RDI: 0000000000000001 +RBP: ffff888033431ac0 R08: 0000000000000001 R09: 0000000000000000 +R10: 0000000000000001 R11: 0000000000000001 R12: ffff888033431a00 +R13: ffff888033431ae4 R14: ffff888033184724 R15: dffffc0000000000 +FS: 0000555580328500(0000) GS:ffff888124a62000(0000) knlGS:0000000000000000 +CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 +CR2: 00007f71733743e0 CR3: 000000007f618000 CR4: 00000000003526f0 +DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 +DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 +Call Trace: + + sunrpc_exit_net+0x46/0x90 net/sunrpc/sunrpc_syms.c:76 + ops_exit_list net/core/net_namespace.c:200 [inline] + ops_undo_list+0x2eb/0xab0 net/core/net_namespace.c:253 + setup_net+0x2e1/0x510 net/core/net_namespace.c:457 + copy_net_ns+0x2a6/0x5f0 net/core/net_namespace.c:574 + create_new_namespaces+0x3ea/0xa90 kernel/nsproxy.c:110 + unshare_nsproxy_namespaces+0xc0/0x1f0 kernel/nsproxy.c:218 + ksys_unshare+0x45b/0xa40 kernel/fork.c:3121 + __do_sys_unshare kernel/fork.c:3192 [inline] + __se_sys_unshare kernel/fork.c:3190 [inline] + __x64_sys_unshare+0x31/0x40 kernel/fork.c:3190 + do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline] + do_syscall_64+0xcd/0x490 arch/x86/entry/syscall_64.c:94 + entry_SYSCALL_64_after_hwframe+0x77/0x7f +RIP: 0033:0x7fa1a6b8e929 +Code: ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 40 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 a8 ff ff ff f7 d8 64 89 01 48 +RSP: 002b:00007fff3a090368 EFLAGS: 00000246 ORIG_RAX: 0000000000000110 +RAX: ffffffffffffffda RBX: 00007fa1a6db5fa0 RCX: 00007fa1a6b8e929 +RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000040000080 +RBP: 00007fa1a6c10b39 R08: 0000000000000000 R09: 0000000000000000 +R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000 +R13: 00007fa1a6db5fa0 R14: 00007fa1a6db5fa0 R15: 0000000000000001 + + +Fixes: d47151b79e32 ("nfs: expose /proc/net/sunrpc/nfs in net namespaces") +Reported-by: syzbot+a4cc4ac22daa4a71b87c@syzkaller.appspotmail.com +Closes: https://syzkaller.appspot.com/bug?extid=a4cc4ac22daa4a71b87c +Tested-by: syzbot+a4cc4ac22daa4a71b87c@syzkaller.appspotmail.com +Signed-off-by: Kuniyuki Iwashima +Signed-off-by: Anna Schumaker +Signed-off-by: Sasha Levin +--- + fs/nfs/inode.c | 17 ++++++++++++++--- + 1 file changed, 14 insertions(+), 3 deletions(-) + +diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c +index 8ab7868807a7d..a2fa6bc4d74e3 100644 +--- a/fs/nfs/inode.c ++++ b/fs/nfs/inode.c +@@ -2589,15 +2589,26 @@ EXPORT_SYMBOL_GPL(nfs_net_id); + static int nfs_net_init(struct net *net) + { + struct nfs_net *nn = net_generic(net, nfs_net_id); ++ int err; + + nfs_clients_init(net); + + if (!rpc_proc_register(net, &nn->rpcstats)) { +- nfs_clients_exit(net); +- return -ENOMEM; ++ err = -ENOMEM; ++ goto err_proc_rpc; + } + +- return nfs_fs_proc_net_init(net); ++ err = nfs_fs_proc_net_init(net); ++ if (err) ++ goto err_proc_nfs; ++ ++ return 0; ++ ++err_proc_nfs: ++ rpc_proc_unregister(net, "nfs"); ++err_proc_rpc: ++ nfs_clients_exit(net); ++ return err; + } + + static void nfs_net_exit(struct net *net) +-- +2.39.5 + diff --git a/queue-6.15/nfsv4-flexfiles-fix-handling-of-nfs-level-errors-in-.patch b/queue-6.15/nfsv4-flexfiles-fix-handling-of-nfs-level-errors-in-.patch new file mode 100644 index 0000000000..7ed4543676 --- /dev/null +++ b/queue-6.15/nfsv4-flexfiles-fix-handling-of-nfs-level-errors-in-.patch @@ -0,0 +1,250 @@ +From cbd7be1d9ad1dcfe63eadb42e8f1238ed632128e Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Thu, 19 Jun 2025 15:16:11 -0400 +Subject: NFSv4/flexfiles: Fix handling of NFS level errors in I/O + +From: Trond Myklebust + +[ Upstream commit 38074de35b015df5623f524d6f2b49a0cd395c40 ] + +Allow the flexfiles error handling to recognise NFS level errors (as +opposed to RPC level errors) and handle them separately. The main +motivator is the NFSERR_PERM errors that get returned if the NFS client +connects to the data server through a port number that is lower than +1024. In that case, the client should disconnect and retry a READ on a +different data server, or it should retry a WRITE after reconnecting. + +Reviewed-by: Tigran Mkrtchyan +Fixes: d67ae825a59d ("pnfs/flexfiles: Add the FlexFile Layout Driver") +Signed-off-by: Trond Myklebust +Signed-off-by: Anna Schumaker +Signed-off-by: Sasha Levin +--- + fs/nfs/flexfilelayout/flexfilelayout.c | 118 ++++++++++++++++++------- + 1 file changed, 84 insertions(+), 34 deletions(-) + +diff --git a/fs/nfs/flexfilelayout/flexfilelayout.c b/fs/nfs/flexfilelayout/flexfilelayout.c +index df48074605967..4bea008dbebd7 100644 +--- a/fs/nfs/flexfilelayout/flexfilelayout.c ++++ b/fs/nfs/flexfilelayout/flexfilelayout.c +@@ -1105,6 +1105,7 @@ static void ff_layout_reset_read(struct nfs_pgio_header *hdr) + } + + static int ff_layout_async_handle_error_v4(struct rpc_task *task, ++ u32 op_status, + struct nfs4_state *state, + struct nfs_client *clp, + struct pnfs_layout_segment *lseg, +@@ -1115,34 +1116,42 @@ static int ff_layout_async_handle_error_v4(struct rpc_task *task, + struct nfs4_deviceid_node *devid = FF_LAYOUT_DEVID_NODE(lseg, idx); + struct nfs4_slot_table *tbl = &clp->cl_session->fc_slot_table; + +- switch (task->tk_status) { +- case -NFS4ERR_BADSESSION: +- case -NFS4ERR_BADSLOT: +- case -NFS4ERR_BAD_HIGH_SLOT: +- case -NFS4ERR_DEADSESSION: +- case -NFS4ERR_CONN_NOT_BOUND_TO_SESSION: +- case -NFS4ERR_SEQ_FALSE_RETRY: +- case -NFS4ERR_SEQ_MISORDERED: ++ switch (op_status) { ++ case NFS4_OK: ++ case NFS4ERR_NXIO: ++ break; ++ case NFSERR_PERM: ++ if (!task->tk_xprt) ++ break; ++ xprt_force_disconnect(task->tk_xprt); ++ goto out_retry; ++ case NFS4ERR_BADSESSION: ++ case NFS4ERR_BADSLOT: ++ case NFS4ERR_BAD_HIGH_SLOT: ++ case NFS4ERR_DEADSESSION: ++ case NFS4ERR_CONN_NOT_BOUND_TO_SESSION: ++ case NFS4ERR_SEQ_FALSE_RETRY: ++ case NFS4ERR_SEQ_MISORDERED: + dprintk("%s ERROR %d, Reset session. Exchangeid " + "flags 0x%x\n", __func__, task->tk_status, + clp->cl_exchange_flags); + nfs4_schedule_session_recovery(clp->cl_session, task->tk_status); +- break; +- case -NFS4ERR_DELAY: ++ goto out_retry; ++ case NFS4ERR_DELAY: + nfs_inc_stats(lseg->pls_layout->plh_inode, NFSIOS_DELAY); + fallthrough; +- case -NFS4ERR_GRACE: ++ case NFS4ERR_GRACE: + rpc_delay(task, FF_LAYOUT_POLL_RETRY_MAX); +- break; +- case -NFS4ERR_RETRY_UNCACHED_REP: +- break; ++ goto out_retry; ++ case NFS4ERR_RETRY_UNCACHED_REP: ++ goto out_retry; + /* Invalidate Layout errors */ +- case -NFS4ERR_PNFS_NO_LAYOUT: +- case -ESTALE: /* mapped NFS4ERR_STALE */ +- case -EBADHANDLE: /* mapped NFS4ERR_BADHANDLE */ +- case -EISDIR: /* mapped NFS4ERR_ISDIR */ +- case -NFS4ERR_FHEXPIRED: +- case -NFS4ERR_WRONG_TYPE: ++ case NFS4ERR_PNFS_NO_LAYOUT: ++ case NFS4ERR_STALE: ++ case NFS4ERR_BADHANDLE: ++ case NFS4ERR_ISDIR: ++ case NFS4ERR_FHEXPIRED: ++ case NFS4ERR_WRONG_TYPE: + dprintk("%s Invalid layout error %d\n", __func__, + task->tk_status); + /* +@@ -1155,6 +1164,11 @@ static int ff_layout_async_handle_error_v4(struct rpc_task *task, + pnfs_destroy_layout(NFS_I(inode)); + rpc_wake_up(&tbl->slot_tbl_waitq); + goto reset; ++ default: ++ break; ++ } ++ ++ switch (task->tk_status) { + /* RPC connection errors */ + case -ENETDOWN: + case -ENETUNREACH: +@@ -1174,27 +1188,56 @@ static int ff_layout_async_handle_error_v4(struct rpc_task *task, + nfs4_delete_deviceid(devid->ld, devid->nfs_client, + &devid->deviceid); + rpc_wake_up(&tbl->slot_tbl_waitq); +- fallthrough; ++ break; + default: +- if (ff_layout_avoid_mds_available_ds(lseg)) +- return -NFS4ERR_RESET_TO_PNFS; +-reset: +- dprintk("%s Retry through MDS. Error %d\n", __func__, +- task->tk_status); +- return -NFS4ERR_RESET_TO_MDS; ++ break; + } ++ ++ if (ff_layout_avoid_mds_available_ds(lseg)) ++ return -NFS4ERR_RESET_TO_PNFS; ++reset: ++ dprintk("%s Retry through MDS. Error %d\n", __func__, ++ task->tk_status); ++ return -NFS4ERR_RESET_TO_MDS; ++ ++out_retry: + task->tk_status = 0; + return -EAGAIN; + } + + /* Retry all errors through either pNFS or MDS except for -EJUKEBOX */ + static int ff_layout_async_handle_error_v3(struct rpc_task *task, ++ u32 op_status, + struct nfs_client *clp, + struct pnfs_layout_segment *lseg, + u32 idx) + { + struct nfs4_deviceid_node *devid = FF_LAYOUT_DEVID_NODE(lseg, idx); + ++ switch (op_status) { ++ case NFS_OK: ++ case NFSERR_NXIO: ++ break; ++ case NFSERR_PERM: ++ if (!task->tk_xprt) ++ break; ++ xprt_force_disconnect(task->tk_xprt); ++ goto out_retry; ++ case NFSERR_ACCES: ++ case NFSERR_BADHANDLE: ++ case NFSERR_FBIG: ++ case NFSERR_IO: ++ case NFSERR_NOSPC: ++ case NFSERR_ROFS: ++ case NFSERR_STALE: ++ goto out_reset_to_pnfs; ++ case NFSERR_JUKEBOX: ++ nfs_inc_stats(lseg->pls_layout->plh_inode, NFSIOS_DELAY); ++ goto out_retry; ++ default: ++ break; ++ } ++ + switch (task->tk_status) { + /* File access problems. Don't mark the device as unavailable */ + case -EACCES: +@@ -1218,6 +1261,7 @@ static int ff_layout_async_handle_error_v3(struct rpc_task *task, + nfs4_delete_deviceid(devid->ld, devid->nfs_client, + &devid->deviceid); + } ++out_reset_to_pnfs: + /* FIXME: Need to prevent infinite looping here. */ + return -NFS4ERR_RESET_TO_PNFS; + out_retry: +@@ -1228,6 +1272,7 @@ static int ff_layout_async_handle_error_v3(struct rpc_task *task, + } + + static int ff_layout_async_handle_error(struct rpc_task *task, ++ u32 op_status, + struct nfs4_state *state, + struct nfs_client *clp, + struct pnfs_layout_segment *lseg, +@@ -1246,10 +1291,11 @@ static int ff_layout_async_handle_error(struct rpc_task *task, + + switch (vers) { + case 3: +- return ff_layout_async_handle_error_v3(task, clp, lseg, idx); +- case 4: +- return ff_layout_async_handle_error_v4(task, state, clp, ++ return ff_layout_async_handle_error_v3(task, op_status, clp, + lseg, idx); ++ case 4: ++ return ff_layout_async_handle_error_v4(task, op_status, state, ++ clp, lseg, idx); + default: + /* should never happen */ + WARN_ON_ONCE(1); +@@ -1302,6 +1348,7 @@ static void ff_layout_io_track_ds_error(struct pnfs_layout_segment *lseg, + switch (status) { + case NFS4ERR_DELAY: + case NFS4ERR_GRACE: ++ case NFS4ERR_PERM: + break; + case NFS4ERR_NXIO: + ff_layout_mark_ds_unreachable(lseg, idx); +@@ -1334,7 +1381,8 @@ static int ff_layout_read_done_cb(struct rpc_task *task, + trace_ff_layout_read_error(hdr, task->tk_status); + } + +- err = ff_layout_async_handle_error(task, hdr->args.context->state, ++ err = ff_layout_async_handle_error(task, hdr->res.op_status, ++ hdr->args.context->state, + hdr->ds_clp, hdr->lseg, + hdr->pgio_mirror_idx); + +@@ -1507,7 +1555,8 @@ static int ff_layout_write_done_cb(struct rpc_task *task, + trace_ff_layout_write_error(hdr, task->tk_status); + } + +- err = ff_layout_async_handle_error(task, hdr->args.context->state, ++ err = ff_layout_async_handle_error(task, hdr->res.op_status, ++ hdr->args.context->state, + hdr->ds_clp, hdr->lseg, + hdr->pgio_mirror_idx); + +@@ -1556,8 +1605,9 @@ static int ff_layout_commit_done_cb(struct rpc_task *task, + trace_ff_layout_commit_error(data, task->tk_status); + } + +- err = ff_layout_async_handle_error(task, NULL, data->ds_clp, +- data->lseg, data->ds_commit_index); ++ err = ff_layout_async_handle_error(task, data->res.op_status, ++ NULL, data->ds_clp, data->lseg, ++ data->ds_commit_index); + + trace_nfs4_pnfs_commit_ds(data, err); + switch (err) { +-- +2.39.5 + diff --git a/queue-6.15/nfsv4-pnfs-fix-a-race-to-wake-on-nfs_layout_drain.patch b/queue-6.15/nfsv4-pnfs-fix-a-race-to-wake-on-nfs_layout_drain.patch new file mode 100644 index 0000000000..4aefc8b6f5 --- /dev/null +++ b/queue-6.15/nfsv4-pnfs-fix-a-race-to-wake-on-nfs_layout_drain.patch @@ -0,0 +1,45 @@ +From 7d3d134b1af4a1c5d01d665711c06136741a0570 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Thu, 19 Jun 2025 11:02:21 -0400 +Subject: NFSv4/pNFS: Fix a race to wake on NFS_LAYOUT_DRAIN + +From: Benjamin Coddington + +[ Upstream commit c01776287414ca43412d1319d2877cbad65444ac ] + +We found a few different systems hung up in writeback waiting on the same +page lock, and one task waiting on the NFS_LAYOUT_DRAIN bit in +pnfs_update_layout(), however the pnfs_layout_hdr's plh_outstanding count +was zero. + +It seems most likely that this is another race between the waiter and waker +similar to commit ed0172af5d6f ("SUNRPC: Fix a race to wake a sync task"). +Fix it up by applying the advised barrier. + +Fixes: 880265c77ac4 ("pNFS: Avoid a live lock condition in pnfs_update_layout()") +Signed-off-by: Benjamin Coddington +Signed-off-by: Anna Schumaker +Signed-off-by: Sasha Levin +--- + fs/nfs/pnfs.c | 4 +++- + 1 file changed, 3 insertions(+), 1 deletion(-) + +diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c +index 3adb7d0dbec7a..1a7ec68bde153 100644 +--- a/fs/nfs/pnfs.c ++++ b/fs/nfs/pnfs.c +@@ -2059,8 +2059,10 @@ static void nfs_layoutget_begin(struct pnfs_layout_hdr *lo) + static void nfs_layoutget_end(struct pnfs_layout_hdr *lo) + { + if (atomic_dec_and_test(&lo->plh_outstanding) && +- test_and_clear_bit(NFS_LAYOUT_DRAIN, &lo->plh_flags)) ++ test_and_clear_bit(NFS_LAYOUT_DRAIN, &lo->plh_flags)) { ++ smp_mb__after_atomic(); + wake_up_bit(&lo->plh_flags, NFS_LAYOUT_DRAIN); ++ } + } + + static bool pnfs_is_first_layoutget(struct pnfs_layout_hdr *lo) +-- +2.39.5 + diff --git a/queue-6.15/nui-fix-dma_mapping_error-check.patch b/queue-6.15/nui-fix-dma_mapping_error-check.patch new file mode 100644 index 0000000000..bd4756d573 --- /dev/null +++ b/queue-6.15/nui-fix-dma_mapping_error-check.patch @@ -0,0 +1,143 @@ +From 322413ff55c4124453a3601e1b51d38086c539bc Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Mon, 30 Jun 2025 10:36:43 +0200 +Subject: nui: Fix dma_mapping_error() check + +From: Thomas Fourier + +[ Upstream commit 561aa0e22b70a5e7246b73d62a824b3aef3fc375 ] + +dma_map_XXX() functions return values DMA_MAPPING_ERROR as error values +which is often ~0. The error value should be tested with +dma_mapping_error(). + +This patch creates a new function in niu_ops to test if the mapping +failed. The test is fixed in niu_rbr_add_page(), added in +niu_start_xmit() and the successfully mapped pages are unmaped upon error. + +Fixes: ec2deec1f352 ("niu: Fix to check for dma mapping errors.") +Signed-off-by: Thomas Fourier +Reviewed-by: Simon Horman +Signed-off-by: David S. Miller +Signed-off-by: Sasha Levin +--- + drivers/net/ethernet/sun/niu.c | 31 ++++++++++++++++++++++++++++++- + drivers/net/ethernet/sun/niu.h | 4 ++++ + 2 files changed, 34 insertions(+), 1 deletion(-) + +diff --git a/drivers/net/ethernet/sun/niu.c b/drivers/net/ethernet/sun/niu.c +index 379b6e90121d9..02b8a9e5c9a90 100644 +--- a/drivers/net/ethernet/sun/niu.c ++++ b/drivers/net/ethernet/sun/niu.c +@@ -3336,7 +3336,7 @@ static int niu_rbr_add_page(struct niu *np, struct rx_ring_info *rp, + + addr = np->ops->map_page(np->device, page, 0, + PAGE_SIZE, DMA_FROM_DEVICE); +- if (!addr) { ++ if (np->ops->mapping_error(np->device, addr)) { + __free_page(page); + return -ENOMEM; + } +@@ -6676,6 +6676,8 @@ static netdev_tx_t niu_start_xmit(struct sk_buff *skb, + len = skb_headlen(skb); + mapping = np->ops->map_single(np->device, skb->data, + len, DMA_TO_DEVICE); ++ if (np->ops->mapping_error(np->device, mapping)) ++ goto out_drop; + + prod = rp->prod; + +@@ -6717,6 +6719,8 @@ static netdev_tx_t niu_start_xmit(struct sk_buff *skb, + mapping = np->ops->map_page(np->device, skb_frag_page(frag), + skb_frag_off(frag), len, + DMA_TO_DEVICE); ++ if (np->ops->mapping_error(np->device, mapping)) ++ goto out_unmap; + + rp->tx_buffs[prod].skb = NULL; + rp->tx_buffs[prod].mapping = mapping; +@@ -6741,6 +6745,19 @@ static netdev_tx_t niu_start_xmit(struct sk_buff *skb, + out: + return NETDEV_TX_OK; + ++out_unmap: ++ while (i--) { ++ const skb_frag_t *frag; ++ ++ prod = PREVIOUS_TX(rp, prod); ++ frag = &skb_shinfo(skb)->frags[i]; ++ np->ops->unmap_page(np->device, rp->tx_buffs[prod].mapping, ++ skb_frag_size(frag), DMA_TO_DEVICE); ++ } ++ ++ np->ops->unmap_single(np->device, rp->tx_buffs[rp->prod].mapping, ++ skb_headlen(skb), DMA_TO_DEVICE); ++ + out_drop: + rp->tx_errors++; + kfree_skb(skb); +@@ -9644,6 +9661,11 @@ static void niu_pci_unmap_single(struct device *dev, u64 dma_address, + dma_unmap_single(dev, dma_address, size, direction); + } + ++static int niu_pci_mapping_error(struct device *dev, u64 addr) ++{ ++ return dma_mapping_error(dev, addr); ++} ++ + static const struct niu_ops niu_pci_ops = { + .alloc_coherent = niu_pci_alloc_coherent, + .free_coherent = niu_pci_free_coherent, +@@ -9651,6 +9673,7 @@ static const struct niu_ops niu_pci_ops = { + .unmap_page = niu_pci_unmap_page, + .map_single = niu_pci_map_single, + .unmap_single = niu_pci_unmap_single, ++ .mapping_error = niu_pci_mapping_error, + }; + + static void niu_driver_version(void) +@@ -10019,6 +10042,11 @@ static void niu_phys_unmap_single(struct device *dev, u64 dma_address, + /* Nothing to do. */ + } + ++static int niu_phys_mapping_error(struct device *dev, u64 dma_address) ++{ ++ return false; ++} ++ + static const struct niu_ops niu_phys_ops = { + .alloc_coherent = niu_phys_alloc_coherent, + .free_coherent = niu_phys_free_coherent, +@@ -10026,6 +10054,7 @@ static const struct niu_ops niu_phys_ops = { + .unmap_page = niu_phys_unmap_page, + .map_single = niu_phys_map_single, + .unmap_single = niu_phys_unmap_single, ++ .mapping_error = niu_phys_mapping_error, + }; + + static int niu_of_probe(struct platform_device *op) +diff --git a/drivers/net/ethernet/sun/niu.h b/drivers/net/ethernet/sun/niu.h +index 04c215f91fc08..0b169c08b0f2d 100644 +--- a/drivers/net/ethernet/sun/niu.h ++++ b/drivers/net/ethernet/sun/niu.h +@@ -2879,6 +2879,9 @@ struct tx_ring_info { + #define NEXT_TX(tp, index) \ + (((index) + 1) < (tp)->pending ? ((index) + 1) : 0) + ++#define PREVIOUS_TX(tp, index) \ ++ (((index) - 1) >= 0 ? ((index) - 1) : (((tp)->pending) - 1)) ++ + static inline u32 niu_tx_avail(struct tx_ring_info *tp) + { + return (tp->pending - +@@ -3140,6 +3143,7 @@ struct niu_ops { + enum dma_data_direction direction); + void (*unmap_single)(struct device *dev, u64 dma_address, + size_t size, enum dma_data_direction direction); ++ int (*mapping_error)(struct device *dev, u64 dma_address); + }; + + struct niu_link_config { +-- +2.39.5 + diff --git a/queue-6.15/nvme-fix-incorrect-cdw15-value-in-passthru-error-log.patch b/queue-6.15/nvme-fix-incorrect-cdw15-value-in-passthru-error-log.patch new file mode 100644 index 0000000000..d49adad977 --- /dev/null +++ b/queue-6.15/nvme-fix-incorrect-cdw15-value-in-passthru-error-log.patch @@ -0,0 +1,38 @@ +From ad2e4996940bdd503e233adcabf0f8f57f5b371b Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Sat, 28 Jun 2025 11:12:32 -0700 +Subject: nvme: Fix incorrect cdw15 value in passthru error logging + +From: Alok Tiwari + +[ Upstream commit 2e96d2d8c2a7a6c2cef45593c028d9c5ef180316 ] + +Fix an error in nvme_log_err_passthru() where cdw14 was incorrectly +printed twice instead of cdw15. This fix ensures accurate logging of +the full passthrough command payload. + +Fixes: 9f079dda1433 ("nvme: allow passthru cmd error logging") +Signed-off-by: Alok Tiwari +Reviewed-by: Martin K. Petersen +Signed-off-by: Christoph Hellwig +Signed-off-by: Sasha Levin +--- + drivers/nvme/host/core.c | 2 +- + 1 file changed, 1 insertion(+), 1 deletion(-) + +diff --git a/drivers/nvme/host/core.c b/drivers/nvme/host/core.c +index d253b82901110..ae584c97f5284 100644 +--- a/drivers/nvme/host/core.c ++++ b/drivers/nvme/host/core.c +@@ -382,7 +382,7 @@ static void nvme_log_err_passthru(struct request *req) + nr->cmd->common.cdw12, + nr->cmd->common.cdw13, + nr->cmd->common.cdw14, +- nr->cmd->common.cdw14); ++ nr->cmd->common.cdw15); + } + + enum nvme_disposition { +-- +2.39.5 + diff --git a/queue-6.15/nvme-multipath-fix-suspicious-rcu-usage-warning.patch b/queue-6.15/nvme-multipath-fix-suspicious-rcu-usage-warning.patch new file mode 100644 index 0000000000..7e3ca547a2 --- /dev/null +++ b/queue-6.15/nvme-multipath-fix-suspicious-rcu-usage-warning.patch @@ -0,0 +1,111 @@ +From 5438b5c5debd3b4c397764fab07200673913172a Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Mon, 30 Jun 2025 15:48:32 +0800 +Subject: nvme-multipath: fix suspicious RCU usage warning + +From: Geliang Tang + +[ Upstream commit d6811074203b13f715ce2480ac64c5b1c773f2a5 ] + +When I run the NVME over TCP test in virtme-ng, I get the following +"suspicious RCU usage" warning in nvme_mpath_add_sysfs_link(): + +''' +[ 5.024557][ T44] nvmet: Created nvm controller 1 for subsystem nqn.2025-06.org.nvmexpress.mptcp for NQN nqn.2014-08.org.nvmexpress:uuid:f7f6b5e0-ff97-4894-98ac-c85309e0bc77. +[ 5.027401][ T183] nvme nvme0: creating 2 I/O queues. +[ 5.029017][ T183] nvme nvme0: mapped 2/0/0 default/read/poll queues. +[ 5.032587][ T183] nvme nvme0: new ctrl: NQN "nqn.2025-06.org.nvmexpress.mptcp", addr 127.0.0.1:4420, hostnqn: nqn.2014-08.org.nvmexpress:uuid:f7f6b5e0-ff97-4894-98ac-c85309e0bc77 +[ 5.042214][ T25] +[ 5.042440][ T25] ============================= +[ 5.042579][ T25] WARNING: suspicious RCU usage +[ 5.042705][ T25] 6.16.0-rc3+ #23 Not tainted +[ 5.042812][ T25] ----------------------------- +[ 5.042934][ T25] drivers/nvme/host/multipath.c:1203 RCU-list traversed in non-reader section!! +[ 5.043111][ T25] +[ 5.043111][ T25] other info that might help us debug this: +[ 5.043111][ T25] +[ 5.043341][ T25] +[ 5.043341][ T25] rcu_scheduler_active = 2, debug_locks = 1 +[ 5.043502][ T25] 3 locks held by kworker/u9:0/25: +[ 5.043615][ T25] #0: ffff888008730948 ((wq_completion)async){+.+.}-{0:0}, at: process_one_work+0x7ed/0x1350 +[ 5.043830][ T25] #1: ffffc900001afd40 ((work_completion)(&entry->work)){+.+.}-{0:0}, at: process_one_work+0xcf3/0x1350 +[ 5.044084][ T25] #2: ffff888013ee0020 (&head->srcu){.+.+}-{0:0}, at: nvme_mpath_add_sysfs_link.part.0+0xb4/0x3a0 +[ 5.044300][ T25] +[ 5.044300][ T25] stack backtrace: +[ 5.044439][ T25] CPU: 0 UID: 0 PID: 25 Comm: kworker/u9:0 Not tainted 6.16.0-rc3+ #23 PREEMPT(full) +[ 5.044441][ T25] Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011 +[ 5.044442][ T25] Workqueue: async async_run_entry_fn +[ 5.044445][ T25] Call Trace: +[ 5.044446][ T25] +[ 5.044449][ T25] dump_stack_lvl+0x6f/0xb0 +[ 5.044453][ T25] lockdep_rcu_suspicious.cold+0x4f/0xb1 +[ 5.044457][ T25] nvme_mpath_add_sysfs_link.part.0+0x2fb/0x3a0 +[ 5.044459][ T25] ? queue_work_on+0x90/0xf0 +[ 5.044461][ T25] ? lockdep_hardirqs_on+0x78/0x110 +[ 5.044466][ T25] nvme_mpath_set_live+0x1e9/0x4f0 +[ 5.044470][ T25] nvme_mpath_add_disk+0x240/0x2f0 +[ 5.044472][ T25] ? __pfx_nvme_mpath_add_disk+0x10/0x10 +[ 5.044475][ T25] ? add_disk_fwnode+0x361/0x580 +[ 5.044480][ T25] nvme_alloc_ns+0x81c/0x17c0 +[ 5.044483][ T25] ? kasan_quarantine_put+0x104/0x240 +[ 5.044487][ T25] ? __pfx_nvme_alloc_ns+0x10/0x10 +[ 5.044495][ T25] ? __pfx_nvme_find_get_ns+0x10/0x10 +[ 5.044496][ T25] ? rcu_read_lock_any_held+0x45/0xa0 +[ 5.044498][ T25] ? validate_chain+0x232/0x4f0 +[ 5.044503][ T25] nvme_scan_ns+0x4c8/0x810 +[ 5.044506][ T25] ? __pfx_nvme_scan_ns+0x10/0x10 +[ 5.044508][ T25] ? find_held_lock+0x2b/0x80 +[ 5.044512][ T25] ? ktime_get+0x16d/0x220 +[ 5.044517][ T25] ? kvm_clock_get_cycles+0x18/0x30 +[ 5.044520][ T25] ? __pfx_nvme_scan_ns_async+0x10/0x10 +[ 5.044522][ T25] async_run_entry_fn+0x97/0x560 +[ 5.044523][ T25] ? rcu_is_watching+0x12/0xc0 +[ 5.044526][ T25] process_one_work+0xd3c/0x1350 +[ 5.044532][ T25] ? __pfx_process_one_work+0x10/0x10 +[ 5.044536][ T25] ? assign_work+0x16c/0x240 +[ 5.044539][ T25] worker_thread+0x4da/0xd50 +[ 5.044545][ T25] ? __pfx_worker_thread+0x10/0x10 +[ 5.044546][ T25] kthread+0x356/0x5c0 +[ 5.044548][ T25] ? __pfx_kthread+0x10/0x10 +[ 5.044549][ T25] ? ret_from_fork+0x1b/0x2e0 +[ 5.044552][ T25] ? __lock_release.isra.0+0x5d/0x180 +[ 5.044553][ T25] ? ret_from_fork+0x1b/0x2e0 +[ 5.044555][ T25] ? rcu_is_watching+0x12/0xc0 +[ 5.044557][ T25] ? __pfx_kthread+0x10/0x10 +[ 5.044559][ T25] ret_from_fork+0x218/0x2e0 +[ 5.044561][ T25] ? __pfx_kthread+0x10/0x10 +[ 5.044562][ T25] ret_from_fork_asm+0x1a/0x30 +[ 5.044570][ T25] +''' + +This patch uses sleepable RCU version of helper list_for_each_entry_srcu() +instead of list_for_each_entry_rcu() to fix it. + +Fixes: 4dbd2b2ebe4c ("nvme-multipath: Add visibility for round-robin io-policy") +Signed-off-by: Geliang Tang +Reviewed-by: Keith Busch +Reviewed-by: Hannes Reinecke +Reviewed-by: Nilay Shroff +Signed-off-by: Christoph Hellwig +Signed-off-by: Sasha Levin +--- + drivers/nvme/host/multipath.c | 3 ++- + 1 file changed, 2 insertions(+), 1 deletion(-) + +diff --git a/drivers/nvme/host/multipath.c b/drivers/nvme/host/multipath.c +index cf0ef4745564c..700dfbd5a4512 100644 +--- a/drivers/nvme/host/multipath.c ++++ b/drivers/nvme/host/multipath.c +@@ -1050,7 +1050,8 @@ void nvme_mpath_add_sysfs_link(struct nvme_ns_head *head) + */ + srcu_idx = srcu_read_lock(&head->srcu); + +- list_for_each_entry_rcu(ns, &head->list, siblings) { ++ list_for_each_entry_srcu(ns, &head->list, siblings, ++ srcu_read_lock_held(&head->srcu)) { + /* + * Ensure that ns path disk node is already added otherwise we + * may get invalid kobj name for target +-- +2.39.5 + diff --git a/queue-6.15/nvme-pci-refresh-visible-attrs-after-being-checked.patch b/queue-6.15/nvme-pci-refresh-visible-attrs-after-being-checked.patch new file mode 100644 index 0000000000..e48e15914b --- /dev/null +++ b/queue-6.15/nvme-pci-refresh-visible-attrs-after-being-checked.patch @@ -0,0 +1,65 @@ +From 8172ae10af7361c206618b76b10d7351fa50305e Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Fri, 13 Jun 2025 16:21:01 -0300 +Subject: nvme-pci: refresh visible attrs after being checked +MIME-Version: 1.0 +Content-Type: text/plain; charset=UTF-8 +Content-Transfer-Encoding: 8bit + +From: Eugen Hristev + +[ Upstream commit 14005c96d6649b27fa52d0cd0f492eb3b5586c07 ] + +The sysfs attributes are registered early, but the driver does not know +whether they are needed or not at that moment. + +For the CMB attributes, commit e917a849c3fc ("nvme-pci: refresh visible +attrs for cmb attributes") solved this problem by +calling nvme_update_attrs after mapping the CMB. However the issue +persists for the HMB attributes. To solve the problem, moved the call to +nvme_update_attrs after nvme_setup_host_mem, which sets up the HMB. + +Fixes: e917a849c3fc ("nvme-pci: refresh visible attrs for cmb attributes") +Fixes: 86adbf0cdb9e ("nvme: simplify transport specific device attribute handling") +Signed-off-by: Eugen Hristev +Signed-off-by: André Almeida +Signed-off-by: Christoph Hellwig +Signed-off-by: Sasha Levin +--- + drivers/nvme/host/pci.c | 6 ++++-- + 1 file changed, 4 insertions(+), 2 deletions(-) + +diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c +index f1dd804151b1c..776c867fb64d5 100644 +--- a/drivers/nvme/host/pci.c ++++ b/drivers/nvme/host/pci.c +@@ -2031,8 +2031,6 @@ static void nvme_map_cmb(struct nvme_dev *dev) + if ((dev->cmbsz & (NVME_CMBSZ_WDS | NVME_CMBSZ_RDS)) == + (NVME_CMBSZ_WDS | NVME_CMBSZ_RDS)) + pci_p2pmem_publish(pdev, true); +- +- nvme_update_attrs(dev); + } + + static int nvme_set_host_mem(struct nvme_dev *dev, u32 bits) +@@ -2969,6 +2967,8 @@ static void nvme_reset_work(struct work_struct *work) + if (result < 0) + goto out; + ++ nvme_update_attrs(dev); ++ + result = nvme_setup_io_queues(dev); + if (result) + goto out; +@@ -3305,6 +3305,8 @@ static int nvme_probe(struct pci_dev *pdev, const struct pci_device_id *id) + if (result < 0) + goto out_disable; + ++ nvme_update_attrs(dev); ++ + result = nvme_setup_io_queues(dev); + if (result) + goto out_disable; +-- +2.39.5 + diff --git a/queue-6.15/nvmet-fix-memory-leak-of-bio-integrity.patch b/queue-6.15/nvmet-fix-memory-leak-of-bio-integrity.patch new file mode 100644 index 0000000000..7ac7b66860 --- /dev/null +++ b/queue-6.15/nvmet-fix-memory-leak-of-bio-integrity.patch @@ -0,0 +1,42 @@ +From 925c37a3e368152b4701889f1807d7eff2d40005 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Wed, 25 Jun 2025 14:45:33 +0300 +Subject: nvmet: fix memory leak of bio integrity + +From: Dmitry Bogdanov + +[ Upstream commit 190f4c2c863af7cc5bb354b70e0805f06419c038 ] + +If nvmet receives commands with metadata there is a continuous memory +leak of kmalloc-128 slab or more precisely bio->bi_integrity. + +Since commit bf4c89fc8797 ("block: don't call bio_uninit from bio_endio") +each user of bio_init has to use bio_uninit as well. Otherwise the bio +integrity is not getting free. Nvmet uses bio_init for inline bios. + +Uninit the inline bio to complete deallocation of integrity in bio. + +Fixes: bf4c89fc8797 ("block: don't call bio_uninit from bio_endio") +Signed-off-by: Dmitry Bogdanov +Signed-off-by: Christoph Hellwig +Signed-off-by: Sasha Levin +--- + drivers/nvme/target/nvmet.h | 2 ++ + 1 file changed, 2 insertions(+) + +diff --git a/drivers/nvme/target/nvmet.h b/drivers/nvme/target/nvmet.h +index b6db8b74dc4ad..85e50e2981742 100644 +--- a/drivers/nvme/target/nvmet.h ++++ b/drivers/nvme/target/nvmet.h +@@ -857,6 +857,8 @@ static inline void nvmet_req_bio_put(struct nvmet_req *req, struct bio *bio) + { + if (bio != &req->b.inline_bio) + bio_put(bio); ++ else ++ bio_uninit(bio); + } + + #ifdef CONFIG_NVME_TARGET_TCP_TLS +-- +2.39.5 + diff --git a/queue-6.15/platform-mellanox-mlxbf-pmc-fix-duplicate-event-id-f.patch b/queue-6.15/platform-mellanox-mlxbf-pmc-fix-duplicate-event-id-f.patch new file mode 100644 index 0000000000..0f7858d0e2 --- /dev/null +++ b/queue-6.15/platform-mellanox-mlxbf-pmc-fix-duplicate-event-id-f.patch @@ -0,0 +1,44 @@ +From 37fc0030d5a39b9ed799611f90ed178c1415673e Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Wed, 18 Jun 2025 23:05:00 -0700 +Subject: platform/mellanox: mlxbf-pmc: Fix duplicate event ID for CACHE_DATA1 +MIME-Version: 1.0 +Content-Type: text/plain; charset=UTF-8 +Content-Transfer-Encoding: 8bit + +From: Alok Tiwari + +[ Upstream commit 173bbec6693f3f3f00dac144f3aa0cd62fb60d33 ] + +same ID (103) was assigned to both GDC_BANK0_G_RSE_PIPE_CACHE_DATA0 +and GDC_BANK0_G_RSE_PIPE_CACHE_DATA1. This could lead to incorrect +event mapping. +Updated the ID to 104 to ensure uniqueness. + +Fixes: 423c3361855c ("platform/mellanox: mlxbf-pmc: Add support for BlueField-3") +Signed-off-by: Alok Tiwari +Reviewed-by: David Thompson +Link: https://lore.kernel.org/r/20250619060502.3594350-1-alok.a.tiwari@oracle.com +Reviewed-by: Ilpo Järvinen +Signed-off-by: Ilpo Järvinen +Signed-off-by: Sasha Levin +--- + drivers/platform/mellanox/mlxbf-pmc.c | 2 +- + 1 file changed, 1 insertion(+), 1 deletion(-) + +diff --git a/drivers/platform/mellanox/mlxbf-pmc.c b/drivers/platform/mellanox/mlxbf-pmc.c +index 36a00692347dc..771a24c397c12 100644 +--- a/drivers/platform/mellanox/mlxbf-pmc.c ++++ b/drivers/platform/mellanox/mlxbf-pmc.c +@@ -713,7 +713,7 @@ static const struct mlxbf_pmc_events mlxbf_pmc_llt_events[] = { + {101, "GDC_BANK0_HIT_DCL_PARTIAL"}, + {102, "GDC_BANK0_EVICT_DCL"}, + {103, "GDC_BANK0_G_RSE_PIPE_CACHE_DATA0"}, +- {103, "GDC_BANK0_G_RSE_PIPE_CACHE_DATA1"}, ++ {104, "GDC_BANK0_G_RSE_PIPE_CACHE_DATA1"}, + {105, "GDC_BANK0_ARB_STRB"}, + {106, "GDC_BANK0_ARB_WAIT"}, + {107, "GDC_BANK0_GGA_STRB"}, +-- +2.39.5 + diff --git a/queue-6.15/platform-mellanox-mlxbf-tmfifo-fix-vring_desc.len-as.patch b/queue-6.15/platform-mellanox-mlxbf-tmfifo-fix-vring_desc.len-as.patch new file mode 100644 index 0000000000..c3a822f9e9 --- /dev/null +++ b/queue-6.15/platform-mellanox-mlxbf-tmfifo-fix-vring_desc.len-as.patch @@ -0,0 +1,46 @@ +From b631089272c9d8693882a659ccb0edcb6aad0430 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Fri, 13 Jun 2025 21:46:08 +0000 +Subject: platform/mellanox: mlxbf-tmfifo: fix vring_desc.len assignment +MIME-Version: 1.0 +Content-Type: text/plain; charset=UTF-8 +Content-Transfer-Encoding: 8bit + +From: David Thompson + +[ Upstream commit 109f4d29dade8ae5b4ac6325af9d1bc24b4230f8 ] + +Fix warnings reported by sparse, related to incorrect type: +drivers/platform/mellanox/mlxbf-tmfifo.c:284:38: warning: incorrect type in assignment (different base types) +drivers/platform/mellanox/mlxbf-tmfifo.c:284:38: expected restricted __virtio32 [usertype] len +drivers/platform/mellanox/mlxbf-tmfifo.c:284:38: got unsigned long + +Reported-by: kernel test robot +Closes: https://lore.kernel.org/oe-kbuild-all/202404040339.S7CUIgf3-lkp@intel.com/ +Fixes: 78034cbece79 ("platform/mellanox: mlxbf-tmfifo: Drop the Rx packet if no more descriptors") +Signed-off-by: David Thompson +Link: https://lore.kernel.org/r/20250613214608.2250130-1-davthompson@nvidia.com +Reviewed-by: Ilpo Järvinen +Signed-off-by: Ilpo Järvinen +Signed-off-by: Sasha Levin +--- + drivers/platform/mellanox/mlxbf-tmfifo.c | 3 ++- + 1 file changed, 2 insertions(+), 1 deletion(-) + +diff --git a/drivers/platform/mellanox/mlxbf-tmfifo.c b/drivers/platform/mellanox/mlxbf-tmfifo.c +index aae99adb29eb0..70c58c4c6c842 100644 +--- a/drivers/platform/mellanox/mlxbf-tmfifo.c ++++ b/drivers/platform/mellanox/mlxbf-tmfifo.c +@@ -281,7 +281,8 @@ static int mlxbf_tmfifo_alloc_vrings(struct mlxbf_tmfifo *fifo, + vring->align = SMP_CACHE_BYTES; + vring->index = i; + vring->vdev_id = tm_vdev->vdev.id.device; +- vring->drop_desc.len = VRING_DROP_DESC_MAX_LEN; ++ vring->drop_desc.len = cpu_to_virtio32(&tm_vdev->vdev, ++ VRING_DROP_DESC_MAX_LEN); + dev = &tm_vdev->vdev.dev; + + size = vring_size(vring->num, vring->align); +-- +2.39.5 + diff --git a/queue-6.15/platform-mellanox-mlxreg-lc-fix-logic-error-in-power.patch b/queue-6.15/platform-mellanox-mlxreg-lc-fix-logic-error-in-power.patch new file mode 100644 index 0000000000..5aeed4f0dc --- /dev/null +++ b/queue-6.15/platform-mellanox-mlxreg-lc-fix-logic-error-in-power.patch @@ -0,0 +1,51 @@ +From 703396c839f371492982f6721d711d38ea37bd8f Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Mon, 30 Jun 2025 03:58:08 -0700 +Subject: platform/mellanox: mlxreg-lc: Fix logic error in power state check +MIME-Version: 1.0 +Content-Type: text/plain; charset=UTF-8 +Content-Transfer-Encoding: 8bit + +From: Alok Tiwari + +[ Upstream commit 644bec18e705ca41d444053407419a21832fcb2f ] + +Fixes a logic issue in mlxreg_lc_completion_notify() where the +intention was to check if MLXREG_LC_POWERED flag is not set before +powering on the device. + +The original code used "state & ~MLXREG_LC_POWERED" to check for the +absence of the POWERED bit. However this condition evaluates to true +even when other bits are set, leading to potentially incorrect +behavior. + +Corrected the logic to explicitly check for the absence of +MLXREG_LC_POWERED using !(state & MLXREG_LC_POWERED). + +Fixes: 62f9529b8d5c ("platform/mellanox: mlxreg-lc: Add initial support for Nvidia line card devices") +Suggested-by: Vadim Pasternak +Signed-off-by: Alok Tiwari +Link: https://lore.kernel.org/r/20250630105812.601014-1-alok.a.tiwari@oracle.com +Reviewed-by: Ilpo Järvinen +Signed-off-by: Ilpo Järvinen +Signed-off-by: Sasha Levin +--- + drivers/platform/mellanox/mlxreg-lc.c | 2 +- + 1 file changed, 1 insertion(+), 1 deletion(-) + +diff --git a/drivers/platform/mellanox/mlxreg-lc.c b/drivers/platform/mellanox/mlxreg-lc.c +index aee395bb48ae4..8681ceb7144ba 100644 +--- a/drivers/platform/mellanox/mlxreg-lc.c ++++ b/drivers/platform/mellanox/mlxreg-lc.c +@@ -688,7 +688,7 @@ static int mlxreg_lc_completion_notify(void *handle, struct i2c_adapter *parent, + if (regval & mlxreg_lc->data->mask) { + mlxreg_lc->state |= MLXREG_LC_SYNCED; + mlxreg_lc_state_update_locked(mlxreg_lc, MLXREG_LC_SYNCED, 1); +- if (mlxreg_lc->state & ~MLXREG_LC_POWERED) { ++ if (!(mlxreg_lc->state & MLXREG_LC_POWERED)) { + err = mlxreg_lc_power_on_off(mlxreg_lc, 1); + if (err) + goto mlxreg_lc_regmap_power_on_off_fail; +-- +2.39.5 + diff --git a/queue-6.15/platform-mellanox-nvsw-sn2201-fix-bus-number-in-adap.patch b/queue-6.15/platform-mellanox-nvsw-sn2201-fix-bus-number-in-adap.patch new file mode 100644 index 0000000000..f74ba41119 --- /dev/null +++ b/queue-6.15/platform-mellanox-nvsw-sn2201-fix-bus-number-in-adap.patch @@ -0,0 +1,42 @@ +From 7188052268e1999499fb07bc192b398402b840d6 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Sun, 22 Jun 2025 00:29:12 -0700 +Subject: platform/mellanox: nvsw-sn2201: Fix bus number in adapter error + message +MIME-Version: 1.0 +Content-Type: text/plain; charset=UTF-8 +Content-Transfer-Encoding: 8bit + +From: Alok Tiwari + +[ Upstream commit d07143b507c51c04c091081627c5a130e9d3c517 ] + +change error log to use correct bus number from main_mux_devs +instead of cpld_devs. + +Fixes: 662f24826f95 ("platform/mellanox: Add support for new SN2201 system") +Signed-off-by: Alok Tiwari +Reviewed-by: Vadim Pasternak +Link: https://lore.kernel.org/r/20250622072921.4111552-2-alok.a.tiwari@oracle.com +Signed-off-by: Ilpo Järvinen +Signed-off-by: Sasha Levin +--- + drivers/platform/mellanox/nvsw-sn2201.c | 2 +- + 1 file changed, 1 insertion(+), 1 deletion(-) + +diff --git a/drivers/platform/mellanox/nvsw-sn2201.c b/drivers/platform/mellanox/nvsw-sn2201.c +index 0c047aa2345b3..490001a9afd76 100644 +--- a/drivers/platform/mellanox/nvsw-sn2201.c ++++ b/drivers/platform/mellanox/nvsw-sn2201.c +@@ -1088,7 +1088,7 @@ static int nvsw_sn2201_i2c_completion_notify(void *handle, int id) + if (!nvsw_sn2201->main_mux_devs->adapter) { + err = -ENODEV; + dev_err(nvsw_sn2201->dev, "Failed to get adapter for bus %d\n", +- nvsw_sn2201->cpld_devs->nr); ++ nvsw_sn2201->main_mux_devs->nr); + goto i2c_get_adapter_main_fail; + } + +-- +2.39.5 + diff --git a/queue-6.15/platform-x86-amd-pmc-add-pcspecialist-lafite-pro-v-1.patch b/queue-6.15/platform-x86-amd-pmc-add-pcspecialist-lafite-pro-v-1.patch new file mode 100644 index 0000000000..9998cc3705 --- /dev/null +++ b/queue-6.15/platform-x86-amd-pmc-add-pcspecialist-lafite-pro-v-1.patch @@ -0,0 +1,56 @@ +From 5015a90fe6f20bc37df8585295f0783f573e42ca Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Wed, 11 Jun 2025 15:33:40 -0500 +Subject: platform/x86/amd/pmc: Add PCSpecialist Lafite Pro V 14M to 8042 + quirks list +MIME-Version: 1.0 +Content-Type: text/plain; charset=UTF-8 +Content-Transfer-Encoding: 8bit + +From: Mario Limonciello + +[ Upstream commit 9ba75ccad85708c5a484637dccc1fc59295b0a83 ] + +Every other s2idle cycle fails to reach hardware sleep when keyboard +wakeup is enabled. This appears to be an EC bug, but the vendor +refuses to fix it. + +It was confirmed that turning off i8042 wakeup avoids ths issue +(albeit keyboard wakeup is disabled). Take the lesser of two evils +and add it to the i8042 quirk list. + +Reported-by: Raoul +Closes: https://bugzilla.kernel.org/show_bug.cgi?id=220116 +Tested-by: Raoul +Signed-off-by: Mario Limonciello +Link: https://lore.kernel.org/r/20250611203341.3733478-1-superm1@kernel.org +Reviewed-by: Ilpo Järvinen +Signed-off-by: Ilpo Järvinen +Signed-off-by: Sasha Levin +--- + drivers/platform/x86/amd/pmc/pmc-quirks.c | 9 +++++++++ + 1 file changed, 9 insertions(+) + +diff --git a/drivers/platform/x86/amd/pmc/pmc-quirks.c b/drivers/platform/x86/amd/pmc/pmc-quirks.c +index 2e3f6fc67c568..7ed12c1d3b34c 100644 +--- a/drivers/platform/x86/amd/pmc/pmc-quirks.c ++++ b/drivers/platform/x86/amd/pmc/pmc-quirks.c +@@ -224,6 +224,15 @@ static const struct dmi_system_id fwbug_list[] = { + DMI_MATCH(DMI_BOARD_NAME, "WUJIE14-GX4HRXL"), + } + }, ++ /* https://bugzilla.kernel.org/show_bug.cgi?id=220116 */ ++ { ++ .ident = "PCSpecialist Lafite Pro V 14M", ++ .driver_data = &quirk_spurious_8042, ++ .matches = { ++ DMI_MATCH(DMI_SYS_VENDOR, "PCSpecialist"), ++ DMI_MATCH(DMI_PRODUCT_NAME, "Lafite Pro V 14M"), ++ } ++ }, + {} + }; + +-- +2.39.5 + diff --git a/queue-6.15/platform-x86-dell-wmi-sysman-fix-class-device-unregi.patch b/queue-6.15/platform-x86-dell-wmi-sysman-fix-class-device-unregi.patch new file mode 100644 index 0000000000..9b71fd1c17 --- /dev/null +++ b/queue-6.15/platform-x86-dell-wmi-sysman-fix-class-device-unregi.patch @@ -0,0 +1,52 @@ +From f51041b4324faf00b834e7e91d0223df916e9425 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Wed, 25 Jun 2025 22:17:37 -0300 +Subject: platform/x86: dell-wmi-sysman: Fix class device unregistration +MIME-Version: 1.0 +Content-Type: text/plain; charset=UTF-8 +Content-Transfer-Encoding: 8bit + +From: Kurt Borja + +[ Upstream commit 314e5ad4782d08858b3abc325c0487bd2abc23a1 ] + +Devices under the firmware_attributes_class do not have unique a dev_t. +Therefore, device_unregister() should be used instead of +device_destroy(), since the latter may match any device with a given +dev_t. + +Fixes: e8a60aa7404b ("platform/x86: Introduce support for Systems Management Driver over WMI for Dell Systems") +Signed-off-by: Kurt Borja +Link: https://lore.kernel.org/r/20250625-dest-fix-v1-3-3a0f342312bb@gmail.com +Reviewed-by: Ilpo Järvinen +Signed-off-by: Ilpo Järvinen +Signed-off-by: Sasha Levin +--- + drivers/platform/x86/dell/dell-wmi-sysman/sysman.c | 4 ++-- + 1 file changed, 2 insertions(+), 2 deletions(-) + +diff --git a/drivers/platform/x86/dell/dell-wmi-sysman/sysman.c b/drivers/platform/x86/dell/dell-wmi-sysman/sysman.c +index 3c74d5e8350a4..f5402b7146572 100644 +--- a/drivers/platform/x86/dell/dell-wmi-sysman/sysman.c ++++ b/drivers/platform/x86/dell/dell-wmi-sysman/sysman.c +@@ -597,7 +597,7 @@ static int __init sysman_init(void) + release_attributes_data(); + + err_destroy_classdev: +- device_destroy(&firmware_attributes_class, MKDEV(0, 0)); ++ device_unregister(wmi_priv.class_dev); + + err_exit_bios_attr_pass_interface: + exit_bios_attr_pass_interface(); +@@ -611,7 +611,7 @@ static int __init sysman_init(void) + static void __exit sysman_exit(void) + { + release_attributes_data(); +- device_destroy(&firmware_attributes_class, MKDEV(0, 0)); ++ device_unregister(wmi_priv.class_dev); + exit_bios_attr_set_interface(); + exit_bios_attr_pass_interface(); + } +-- +2.39.5 + diff --git a/queue-6.15/platform-x86-dell-wmi-sysman-fix-wmi-data-block-retr.patch b/queue-6.15/platform-x86-dell-wmi-sysman-fix-wmi-data-block-retr.patch new file mode 100644 index 0000000000..1ed821ed18 --- /dev/null +++ b/queue-6.15/platform-x86-dell-wmi-sysman-fix-wmi-data-block-retr.patch @@ -0,0 +1,141 @@ +From 764f642c735cf17c2b2e925c2ac4e8e6f8159489 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Mon, 30 Jun 2025 00:43:12 -0300 +Subject: platform/x86: dell-wmi-sysman: Fix WMI data block retrieval in sysfs + callbacks +MIME-Version: 1.0 +Content-Type: text/plain; charset=UTF-8 +Content-Transfer-Encoding: 8bit + +From: Kurt Borja + +[ Upstream commit eb617dd25ca176f3fee24f873f0fd60010773d67 ] + +After retrieving WMI data blocks in sysfs callbacks, check for the +validity of them before dereferencing their content. + +Reported-by: Jan Graczyk +Closes: https://lore.kernel.org/r/CAHk-=wgMiSKXf7SvQrfEnxVtmT=QVQPjJdNjfm3aXS7wc=rzTw@mail.gmail.com/ +Fixes: e8a60aa7404b ("platform/x86: Introduce support for Systems Management Driver over WMI for Dell Systems") +Suggested-by: Linus Torvalds +Reviewed-by: Armin Wolf +Signed-off-by: Kurt Borja +Link: https://lore.kernel.org/r/20250630-sysman-fix-v2-1-d185674d0a30@gmail.com +Reviewed-by: Ilpo Järvinen +Signed-off-by: Ilpo Järvinen +Signed-off-by: Sasha Levin +--- + .../platform/x86/dell/dell-wmi-sysman/dell-wmi-sysman.h | 5 +++++ + .../platform/x86/dell/dell-wmi-sysman/enum-attributes.c | 5 +++-- + .../platform/x86/dell/dell-wmi-sysman/int-attributes.c | 5 +++-- + .../x86/dell/dell-wmi-sysman/passobj-attributes.c | 5 +++-- + .../platform/x86/dell/dell-wmi-sysman/string-attributes.c | 5 +++-- + drivers/platform/x86/dell/dell-wmi-sysman/sysman.c | 8 ++++---- + 6 files changed, 21 insertions(+), 12 deletions(-) + +diff --git a/drivers/platform/x86/dell/dell-wmi-sysman/dell-wmi-sysman.h b/drivers/platform/x86/dell/dell-wmi-sysman/dell-wmi-sysman.h +index 3ad33a094588c..817ee7ba07ca0 100644 +--- a/drivers/platform/x86/dell/dell-wmi-sysman/dell-wmi-sysman.h ++++ b/drivers/platform/x86/dell/dell-wmi-sysman/dell-wmi-sysman.h +@@ -89,6 +89,11 @@ extern struct wmi_sysman_priv wmi_priv; + + enum { ENUM, INT, STR, PO }; + ++#define ENUM_MIN_ELEMENTS 8 ++#define INT_MIN_ELEMENTS 9 ++#define STR_MIN_ELEMENTS 8 ++#define PO_MIN_ELEMENTS 4 ++ + enum { + ATTR_NAME, + DISPL_NAME_LANG_CODE, +diff --git a/drivers/platform/x86/dell/dell-wmi-sysman/enum-attributes.c b/drivers/platform/x86/dell/dell-wmi-sysman/enum-attributes.c +index 8cc212c852668..fc2f58b4cbc6e 100644 +--- a/drivers/platform/x86/dell/dell-wmi-sysman/enum-attributes.c ++++ b/drivers/platform/x86/dell/dell-wmi-sysman/enum-attributes.c +@@ -23,9 +23,10 @@ static ssize_t current_value_show(struct kobject *kobj, struct kobj_attribute *a + obj = get_wmiobj_pointer(instance_id, DELL_WMI_BIOS_ENUMERATION_ATTRIBUTE_GUID); + if (!obj) + return -EIO; +- if (obj->package.elements[CURRENT_VAL].type != ACPI_TYPE_STRING) { ++ if (obj->type != ACPI_TYPE_PACKAGE || obj->package.count < ENUM_MIN_ELEMENTS || ++ obj->package.elements[CURRENT_VAL].type != ACPI_TYPE_STRING) { + kfree(obj); +- return -EINVAL; ++ return -EIO; + } + ret = snprintf(buf, PAGE_SIZE, "%s\n", obj->package.elements[CURRENT_VAL].string.pointer); + kfree(obj); +diff --git a/drivers/platform/x86/dell/dell-wmi-sysman/int-attributes.c b/drivers/platform/x86/dell/dell-wmi-sysman/int-attributes.c +index 951e75b538fad..7352480642391 100644 +--- a/drivers/platform/x86/dell/dell-wmi-sysman/int-attributes.c ++++ b/drivers/platform/x86/dell/dell-wmi-sysman/int-attributes.c +@@ -25,9 +25,10 @@ static ssize_t current_value_show(struct kobject *kobj, struct kobj_attribute *a + obj = get_wmiobj_pointer(instance_id, DELL_WMI_BIOS_INTEGER_ATTRIBUTE_GUID); + if (!obj) + return -EIO; +- if (obj->package.elements[CURRENT_VAL].type != ACPI_TYPE_INTEGER) { ++ if (obj->type != ACPI_TYPE_PACKAGE || obj->package.count < INT_MIN_ELEMENTS || ++ obj->package.elements[CURRENT_VAL].type != ACPI_TYPE_INTEGER) { + kfree(obj); +- return -EINVAL; ++ return -EIO; + } + ret = snprintf(buf, PAGE_SIZE, "%lld\n", obj->package.elements[CURRENT_VAL].integer.value); + kfree(obj); +diff --git a/drivers/platform/x86/dell/dell-wmi-sysman/passobj-attributes.c b/drivers/platform/x86/dell/dell-wmi-sysman/passobj-attributes.c +index d8f1bf5e58a0f..3167e06d416ed 100644 +--- a/drivers/platform/x86/dell/dell-wmi-sysman/passobj-attributes.c ++++ b/drivers/platform/x86/dell/dell-wmi-sysman/passobj-attributes.c +@@ -26,9 +26,10 @@ static ssize_t is_enabled_show(struct kobject *kobj, struct kobj_attribute *attr + obj = get_wmiobj_pointer(instance_id, DELL_WMI_BIOS_PASSOBJ_ATTRIBUTE_GUID); + if (!obj) + return -EIO; +- if (obj->package.elements[IS_PASS_SET].type != ACPI_TYPE_INTEGER) { ++ if (obj->type != ACPI_TYPE_PACKAGE || obj->package.count < PO_MIN_ELEMENTS || ++ obj->package.elements[IS_PASS_SET].type != ACPI_TYPE_INTEGER) { + kfree(obj); +- return -EINVAL; ++ return -EIO; + } + ret = snprintf(buf, PAGE_SIZE, "%lld\n", obj->package.elements[IS_PASS_SET].integer.value); + kfree(obj); +diff --git a/drivers/platform/x86/dell/dell-wmi-sysman/string-attributes.c b/drivers/platform/x86/dell/dell-wmi-sysman/string-attributes.c +index c392f0ecf8b55..0d2c74f8d1aad 100644 +--- a/drivers/platform/x86/dell/dell-wmi-sysman/string-attributes.c ++++ b/drivers/platform/x86/dell/dell-wmi-sysman/string-attributes.c +@@ -25,9 +25,10 @@ static ssize_t current_value_show(struct kobject *kobj, struct kobj_attribute *a + obj = get_wmiobj_pointer(instance_id, DELL_WMI_BIOS_STRING_ATTRIBUTE_GUID); + if (!obj) + return -EIO; +- if (obj->package.elements[CURRENT_VAL].type != ACPI_TYPE_STRING) { ++ if (obj->type != ACPI_TYPE_PACKAGE || obj->package.count < STR_MIN_ELEMENTS || ++ obj->package.elements[CURRENT_VAL].type != ACPI_TYPE_STRING) { + kfree(obj); +- return -EINVAL; ++ return -EIO; + } + ret = snprintf(buf, PAGE_SIZE, "%s\n", obj->package.elements[CURRENT_VAL].string.pointer); + kfree(obj); +diff --git a/drivers/platform/x86/dell/dell-wmi-sysman/sysman.c b/drivers/platform/x86/dell/dell-wmi-sysman/sysman.c +index d00389b860e4e..3c74d5e8350a4 100644 +--- a/drivers/platform/x86/dell/dell-wmi-sysman/sysman.c ++++ b/drivers/platform/x86/dell/dell-wmi-sysman/sysman.c +@@ -407,10 +407,10 @@ static int init_bios_attributes(int attr_type, const char *guid) + return retval; + + switch (attr_type) { +- case ENUM: min_elements = 8; break; +- case INT: min_elements = 9; break; +- case STR: min_elements = 8; break; +- case PO: min_elements = 4; break; ++ case ENUM: min_elements = ENUM_MIN_ELEMENTS; break; ++ case INT: min_elements = INT_MIN_ELEMENTS; break; ++ case STR: min_elements = STR_MIN_ELEMENTS; break; ++ case PO: min_elements = PO_MIN_ELEMENTS; break; + default: + pr_err("Error: Unknown attr_type: %d\n", attr_type); + return -EINVAL; +-- +2.39.5 + diff --git a/queue-6.15/platform-x86-hp-bioscfg-fix-class-device-unregistrat.patch b/queue-6.15/platform-x86-hp-bioscfg-fix-class-device-unregistrat.patch new file mode 100644 index 0000000000..2261a5b7ec --- /dev/null +++ b/queue-6.15/platform-x86-hp-bioscfg-fix-class-device-unregistrat.patch @@ -0,0 +1,52 @@ +From 02e28fc00ee16897254801b4b9f00737f5661c53 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Wed, 25 Jun 2025 22:17:35 -0300 +Subject: platform/x86: hp-bioscfg: Fix class device unregistration +MIME-Version: 1.0 +Content-Type: text/plain; charset=UTF-8 +Content-Transfer-Encoding: 8bit + +From: Kurt Borja + +[ Upstream commit 11cba4793b95df3bc192149a6eb044f69aa0b99e ] + +Devices under the firmware_attributes_class do not have unique a dev_t. +Therefore, device_unregister() should be used instead of +device_destroy(), since the latter may match any device with a given +dev_t. + +Fixes: a34fc329b189 ("platform/x86: hp-bioscfg: bioscfg") +Signed-off-by: Kurt Borja +Link: https://lore.kernel.org/r/20250625-dest-fix-v1-1-3a0f342312bb@gmail.com +Reviewed-by: Ilpo Järvinen +Signed-off-by: Ilpo Järvinen +Signed-off-by: Sasha Levin +--- + drivers/platform/x86/hp/hp-bioscfg/bioscfg.c | 4 ++-- + 1 file changed, 2 insertions(+), 2 deletions(-) + +diff --git a/drivers/platform/x86/hp/hp-bioscfg/bioscfg.c b/drivers/platform/x86/hp/hp-bioscfg/bioscfg.c +index 13237890fc920..5bfa7159f5bcd 100644 +--- a/drivers/platform/x86/hp/hp-bioscfg/bioscfg.c ++++ b/drivers/platform/x86/hp/hp-bioscfg/bioscfg.c +@@ -1034,7 +1034,7 @@ static int __init hp_init(void) + release_attributes_data(); + + err_destroy_classdev: +- device_destroy(&firmware_attributes_class, MKDEV(0, 0)); ++ device_unregister(bioscfg_drv.class_dev); + + err_unregister_class: + hp_exit_attr_set_interface(); +@@ -1045,7 +1045,7 @@ static int __init hp_init(void) + static void __exit hp_exit(void) + { + release_attributes_data(); +- device_destroy(&firmware_attributes_class, MKDEV(0, 0)); ++ device_unregister(bioscfg_drv.class_dev); + + hp_exit_attr_set_interface(); + } +-- +2.39.5 + diff --git a/queue-6.15/platform-x86-think-lmi-fix-class-device-unregistrati.patch b/queue-6.15/platform-x86-think-lmi-fix-class-device-unregistrati.patch new file mode 100644 index 0000000000..17d21e3d9b --- /dev/null +++ b/queue-6.15/platform-x86-think-lmi-fix-class-device-unregistrati.patch @@ -0,0 +1,52 @@ +From c27e2ff6c45299e9e73874977c0e9966faf24577 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Wed, 25 Jun 2025 22:17:36 -0300 +Subject: platform/x86: think-lmi: Fix class device unregistration +MIME-Version: 1.0 +Content-Type: text/plain; charset=UTF-8 +Content-Transfer-Encoding: 8bit + +From: Kurt Borja + +[ Upstream commit 5ff1fbb3059730700b4823f43999fc1315984632 ] + +Devices under the firmware_attributes_class do not have unique a dev_t. +Therefore, device_unregister() should be used instead of +device_destroy(), since the latter may match any device with a given +dev_t. + +Fixes: a40cd7ef22fb ("platform/x86: think-lmi: Add WMI interface support on Lenovo platforms") +Signed-off-by: Kurt Borja +Link: https://lore.kernel.org/r/20250625-dest-fix-v1-2-3a0f342312bb@gmail.com +Reviewed-by: Ilpo Järvinen +Signed-off-by: Ilpo Järvinen +Signed-off-by: Sasha Levin +--- + drivers/platform/x86/think-lmi.c | 4 ++-- + 1 file changed, 2 insertions(+), 2 deletions(-) + +diff --git a/drivers/platform/x86/think-lmi.c b/drivers/platform/x86/think-lmi.c +index 00b1e7c79a3d1..28ce475f6c377 100644 +--- a/drivers/platform/x86/think-lmi.c ++++ b/drivers/platform/x86/think-lmi.c +@@ -1554,7 +1554,7 @@ static int tlmi_sysfs_init(void) + fail_create_attr: + tlmi_release_attr(); + fail_device_created: +- device_destroy(&firmware_attributes_class, MKDEV(0, 0)); ++ device_unregister(tlmi_priv.class_dev); + fail_class_created: + return ret; + } +@@ -1781,7 +1781,7 @@ static int tlmi_analyze(struct wmi_device *wdev) + static void tlmi_remove(struct wmi_device *wdev) + { + tlmi_release_attr(); +- device_destroy(&firmware_attributes_class, MKDEV(0, 0)); ++ device_unregister(tlmi_priv.class_dev); + } + + static int tlmi_probe(struct wmi_device *wdev, const void *context) +-- +2.39.5 + diff --git a/queue-6.15/platform-x86-wmi-fix-wmi-event-enablement.patch b/queue-6.15/platform-x86-wmi-fix-wmi-event-enablement.patch new file mode 100644 index 0000000000..130404310a --- /dev/null +++ b/queue-6.15/platform-x86-wmi-fix-wmi-event-enablement.patch @@ -0,0 +1,74 @@ +From d3710a46906258261d470b3bad5a51ced69afd40 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Fri, 20 Jun 2025 00:14:39 +0200 +Subject: platform/x86: wmi: Fix WMI event enablement +MIME-Version: 1.0 +Content-Type: text/plain; charset=UTF-8 +Content-Transfer-Encoding: 8bit + +From: Armin Wolf + +[ Upstream commit cf0b812500e64a7d5e2957abed38c3a97917b34f ] + +It turns out that the Windows WMI-ACPI driver always enables/disables +WMI events regardless of whether they are marked as expensive or not. +This finding is further reinforced when reading the documentation of +the WMI_FUNCTION_CONTROL_CALLBACK callback used by Windows drivers +for enabling/disabling WMI devices: + + The DpWmiFunctionControl routine enables or disables + notification of events, and enables or disables data + collection for data blocks that the driver registered + as expensive to collect. + +Follow this behavior to fix the WMI event used for reporting hotkey +events on the Dell Latitude 5400 and likely many more devices. + +Reported-by: Dmytro Bagrii +Closes: https://bugzilla.kernel.org/show_bug.cgi?id=220246 +Tested-by: Dmytro Bagrii +Fixes: 656f0961d126 ("platform/x86: wmi: Rework WCxx/WExx ACPI method handling") +Signed-off-by: Armin Wolf +Link: https://lore.kernel.org/r/20250619221440.6737-1-W_Armin@gmx.de +Reviewed-by: Ilpo Järvinen +Signed-off-by: Ilpo Järvinen +Signed-off-by: Sasha Levin +--- + drivers/platform/x86/wmi.c | 16 +++++++++++----- + 1 file changed, 11 insertions(+), 5 deletions(-) + +diff --git a/drivers/platform/x86/wmi.c b/drivers/platform/x86/wmi.c +index e46453750d5f1..03aecf8bb7f8e 100644 +--- a/drivers/platform/x86/wmi.c ++++ b/drivers/platform/x86/wmi.c +@@ -177,16 +177,22 @@ static int wmi_device_enable(struct wmi_device *wdev, bool enable) + acpi_handle handle; + acpi_status status; + +- if (!(wblock->gblock.flags & ACPI_WMI_EXPENSIVE)) +- return 0; +- + if (wblock->dev.dev.type == &wmi_type_method) + return 0; + +- if (wblock->dev.dev.type == &wmi_type_event) ++ if (wblock->dev.dev.type == &wmi_type_event) { ++ /* ++ * Windows always enables/disables WMI events, even when they are ++ * not marked as being expensive. We follow this behavior for ++ * compatibility reasons. ++ */ + snprintf(method, sizeof(method), "WE%02X", wblock->gblock.notify_id); +- else ++ } else { ++ if (!(wblock->gblock.flags & ACPI_WMI_EXPENSIVE)) ++ return 0; ++ + get_acpi_method_name(wblock, 'C', method); ++ } + + /* + * Not all WMI devices marked as expensive actually implement the +-- +2.39.5 + diff --git a/queue-6.15/powerpc-fix-struct-termio-related-ioctl-macros.patch b/queue-6.15/powerpc-fix-struct-termio-related-ioctl-macros.patch new file mode 100644 index 0000000000..2760a22383 --- /dev/null +++ b/queue-6.15/powerpc-fix-struct-termio-related-ioctl-macros.patch @@ -0,0 +1,58 @@ +From df549717903c50eb9de544c092c0c50ba22a99dd Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Sat, 17 May 2025 19:52:37 +0530 +Subject: powerpc: Fix struct termio related ioctl macros + +From: Madhavan Srinivasan + +[ Upstream commit ab107276607af90b13a5994997e19b7b9731e251 ] + +Since termio interface is now obsolete, include/uapi/asm/ioctls.h +has some constant macros referring to "struct termio", this caused +build failure at userspace. + +In file included from /usr/include/asm/ioctl.h:12, + from /usr/include/asm/ioctls.h:5, + from tst-ioctls.c:3: +tst-ioctls.c: In function 'get_TCGETA': +tst-ioctls.c:12:10: error: invalid application of 'sizeof' to incomplete type 'struct termio' + 12 | return TCGETA; + | ^~~~~~ + +Even though termios.h provides "struct termio", trying to juggle definitions around to +make it compile could introduce regressions. So better to open code it. + +Reported-by: Tulio Magno +Suggested-by: Nicholas Piggin +Tested-by: Justin M. Forbes +Reviewed-by: Michael Ellerman +Closes: https://lore.kernel.org/linuxppc-dev/8734dji5wl.fsf@ascii.art.br/ +Signed-off-by: Madhavan Srinivasan +Link: https://patch.msgid.link/20250517142237.156665-1-maddy@linux.ibm.com +Signed-off-by: Sasha Levin +--- + arch/powerpc/include/uapi/asm/ioctls.h | 8 ++++---- + 1 file changed, 4 insertions(+), 4 deletions(-) + +diff --git a/arch/powerpc/include/uapi/asm/ioctls.h b/arch/powerpc/include/uapi/asm/ioctls.h +index 2c145da3b774a..b5211e413829a 100644 +--- a/arch/powerpc/include/uapi/asm/ioctls.h ++++ b/arch/powerpc/include/uapi/asm/ioctls.h +@@ -23,10 +23,10 @@ + #define TCSETSW _IOW('t', 21, struct termios) + #define TCSETSF _IOW('t', 22, struct termios) + +-#define TCGETA _IOR('t', 23, struct termio) +-#define TCSETA _IOW('t', 24, struct termio) +-#define TCSETAW _IOW('t', 25, struct termio) +-#define TCSETAF _IOW('t', 28, struct termio) ++#define TCGETA 0x40147417 /* _IOR('t', 23, struct termio) */ ++#define TCSETA 0x80147418 /* _IOW('t', 24, struct termio) */ ++#define TCSETAW 0x80147419 /* _IOW('t', 25, struct termio) */ ++#define TCSETAF 0x8014741c /* _IOW('t', 28, struct termio) */ + + #define TCSBRK _IO('t', 29) + #define TCXONC _IO('t', 30) +-- +2.39.5 + diff --git a/queue-6.15/rcu-return-early-if-callback-is-not-specified.patch b/queue-6.15/rcu-return-early-if-callback-is-not-specified.patch new file mode 100644 index 0000000000..5bb0e68b2e --- /dev/null +++ b/queue-6.15/rcu-return-early-if-callback-is-not-specified.patch @@ -0,0 +1,43 @@ +From 36dd4ec375d0ccf54c77a2feb5a513e6c452f6db Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Tue, 10 Jun 2025 19:34:48 +0200 +Subject: rcu: Return early if callback is not specified + +From: Uladzislau Rezki (Sony) + +[ Upstream commit 33b6a1f155d627f5bd80c7485c598ce45428f74f ] + +Currently the call_rcu() API does not check whether a callback +pointer is NULL. If NULL is passed, rcu_core() will try to invoke +it, resulting in NULL pointer dereference and a kernel crash. + +To prevent this and improve debuggability, this patch adds a check +for NULL and emits a kernel stack trace to help identify a faulty +caller. + +Signed-off-by: Uladzislau Rezki (Sony) +Reviewed-by: Joel Fernandes +Signed-off-by: Joel Fernandes +Signed-off-by: Sasha Levin +--- + kernel/rcu/tree.c | 4 ++++ + 1 file changed, 4 insertions(+) + +diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c +index 80b10893b5038..37778d913f011 100644 +--- a/kernel/rcu/tree.c ++++ b/kernel/rcu/tree.c +@@ -3068,6 +3068,10 @@ __call_rcu_common(struct rcu_head *head, rcu_callback_t func, bool lazy_in) + /* Misaligned rcu_head! */ + WARN_ON_ONCE((unsigned long)head & (sizeof(void *) - 1)); + ++ /* Avoid NULL dereference if callback is NULL. */ ++ if (WARN_ON_ONCE(!func)) ++ return; ++ + if (debug_rcu_head_queue(head)) { + /* + * Probable double call_rcu(), so leak the callback. +-- +2.39.5 + diff --git a/queue-6.15/rdma-mlx5-fix-cc-counters-query-for-mpv.patch b/queue-6.15/rdma-mlx5-fix-cc-counters-query-for-mpv.patch new file mode 100644 index 0000000000..7b2553c478 --- /dev/null +++ b/queue-6.15/rdma-mlx5-fix-cc-counters-query-for-mpv.patch @@ -0,0 +1,38 @@ +From d17f29c52b837412614ab1a738261dcee5ca92b0 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Mon, 16 Jun 2025 12:14:53 +0300 +Subject: RDMA/mlx5: Fix CC counters query for MPV + +From: Patrisious Haddad + +[ Upstream commit acd245b1e33fc4b9d0f2e3372021d632f7ee0652 ] + +In case, CC counters are querying for the second port use the correct +core device for the query instead of always using the master core device. + +Fixes: aac4492ef23a ("IB/mlx5: Update counter implementation for dual port RoCE") +Signed-off-by: Patrisious Haddad +Reviewed-by: Michael Guralnik +Link: https://patch.msgid.link/9cace74dcf106116118bebfa9146d40d4166c6b0.1750064969.git.leon@kernel.org +Signed-off-by: Leon Romanovsky +Signed-off-by: Sasha Levin +--- + drivers/infiniband/hw/mlx5/counters.c | 2 +- + 1 file changed, 1 insertion(+), 1 deletion(-) + +diff --git a/drivers/infiniband/hw/mlx5/counters.c b/drivers/infiniband/hw/mlx5/counters.c +index 943e9eb2ad20d..a506fafd2b151 100644 +--- a/drivers/infiniband/hw/mlx5/counters.c ++++ b/drivers/infiniband/hw/mlx5/counters.c +@@ -418,7 +418,7 @@ static int do_get_hw_stats(struct ib_device *ibdev, + */ + goto done; + } +- ret = mlx5_lag_query_cong_counters(dev->mdev, ++ ret = mlx5_lag_query_cong_counters(mdev, + stats->value + + cnts->num_q_counters, + cnts->num_cong_counters, +-- +2.39.5 + diff --git a/queue-6.15/rdma-mlx5-fix-hw-counters-query-for-non-representor-.patch b/queue-6.15/rdma-mlx5-fix-hw-counters-query-for-non-representor-.patch new file mode 100644 index 0000000000..a08f8a8d00 --- /dev/null +++ b/queue-6.15/rdma-mlx5-fix-hw-counters-query-for-non-representor-.patch @@ -0,0 +1,46 @@ +From 8c4a8324609544a4cdf779ca338d9e9c5f2eb6cb Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Mon, 16 Jun 2025 12:14:52 +0300 +Subject: RDMA/mlx5: Fix HW counters query for non-representor devices + +From: Patrisious Haddad + +[ Upstream commit 3cc1dbfddf88dc5ecce0a75185061403b1f7352d ] + +To get the device HW counters, a non-representor switchdev device +should use the mlx5_ib_query_q_counters() function and query all of +the available counters. While a representor device in switchdev mode +should use the mlx5_ib_query_q_counters_vport() function and query only +the Q_Counters without the PPCNT counters and congestion control counters, +since they aren't relevant for a representor device. + +Currently a non-representor switchdev device skips querying the PPCNT +counters and congestion control counters, leaving them unupdated. +Fix that by properly querying those counters for non-representor devices. + +Fixes: d22467a71ebe ("RDMA/mlx5: Expand switchdev Q-counters to expose representor statistics") +Signed-off-by: Patrisious Haddad +Reviewed-by: Maher Sanalla +Link: https://patch.msgid.link/56bf8af4ca8c58e3fb9f7e47b1dca2009eeeed81.1750064969.git.leon@kernel.org +Signed-off-by: Leon Romanovsky +Signed-off-by: Sasha Levin +--- + drivers/infiniband/hw/mlx5/counters.c | 2 +- + 1 file changed, 1 insertion(+), 1 deletion(-) + +diff --git a/drivers/infiniband/hw/mlx5/counters.c b/drivers/infiniband/hw/mlx5/counters.c +index b847084dcd998..943e9eb2ad20d 100644 +--- a/drivers/infiniband/hw/mlx5/counters.c ++++ b/drivers/infiniband/hw/mlx5/counters.c +@@ -398,7 +398,7 @@ static int do_get_hw_stats(struct ib_device *ibdev, + return ret; + + /* We don't expose device counters over Vports */ +- if (is_mdev_switchdev_mode(dev->mdev) && port_num != 0) ++ if (is_mdev_switchdev_mode(dev->mdev) && dev->is_rep && port_num != 0) + goto done; + + if (MLX5_CAP_PCAM_FEATURE(dev->mdev, rx_icrc_encapsulated_counter)) { +-- +2.39.5 + diff --git a/queue-6.15/rdma-mlx5-fix-unsafe-xarray-access-in-implicit-odp-h.patch b/queue-6.15/rdma-mlx5-fix-unsafe-xarray-access-in-implicit-odp-h.patch new file mode 100644 index 0000000000..f977d536b8 --- /dev/null +++ b/queue-6.15/rdma-mlx5-fix-unsafe-xarray-access-in-implicit-odp-h.patch @@ -0,0 +1,104 @@ +From 16b9804ab43a3ded3ee778240e5d03a93f956b1c Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Mon, 16 Jun 2025 11:17:01 +0300 +Subject: RDMA/mlx5: Fix unsafe xarray access in implicit ODP handling + +From: Or Har-Toov + +[ Upstream commit 2c6b640ea08bff1a192bf87fa45246ff1e40767c ] + +__xa_store() and __xa_erase() were used without holding the proper lock, +which led to a lockdep warning due to unsafe RCU usage. This patch +replaces them with xa_store() and xa_erase(), which perform the necessary +locking internally. + + ============================= + WARNING: suspicious RCPU usage + 6.14.0-rc7_for_upstream_debug_2025_03_18_15_01 #1 Not tainted + ----------------------------- + ./include/linux/xarray.h:1211 suspicious rcu_dereference_protected() usage! + + other info that might help us debug this: + + rcu_scheduler_active = 2, debug_locks = 1 + 3 locks held by kworker/u136:0/219: + at: process_one_work+0xbe4/0x15f0 + process_one_work+0x75c/0x15f0 + pagefault_mr+0x9a5/0x1390 [mlx5_ib] + + stack backtrace: + CPU: 14 UID: 0 PID: 219 Comm: kworker/u136:0 Not tainted + 6.14.0-rc7_for_upstream_debug_2025_03_18_15_01 #1 + Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS + rel-1.16.0-0-gd239552ce722-prebuilt.qemu.org 04/01/2014 + Workqueue: mlx5_ib_page_fault mlx5_ib_eqe_pf_action [mlx5_ib] + Call Trace: + dump_stack_lvl+0xa8/0xc0 + lockdep_rcu_suspicious+0x1e6/0x260 + xas_create+0xb8a/0xee0 + xas_store+0x73/0x14c0 + __xa_store+0x13c/0x220 + ? xa_store_range+0x390/0x390 + ? spin_bug+0x1d0/0x1d0 + pagefault_mr+0xcb5/0x1390 [mlx5_ib] + ? _raw_spin_unlock+0x1f/0x30 + mlx5_ib_eqe_pf_action+0x3be/0x2620 [mlx5_ib] + ? lockdep_hardirqs_on_prepare+0x400/0x400 + ? mlx5_ib_invalidate_range+0xcb0/0xcb0 [mlx5_ib] + process_one_work+0x7db/0x15f0 + ? pwq_dec_nr_in_flight+0xda0/0xda0 + ? assign_work+0x168/0x240 + worker_thread+0x57d/0xcd0 + ? rescuer_thread+0xc40/0xc40 + kthread+0x3b3/0x800 + ? kthread_is_per_cpu+0xb0/0xb0 + ? lock_downgrade+0x680/0x680 + ? do_raw_spin_lock+0x12d/0x270 + ? spin_bug+0x1d0/0x1d0 + ? finish_task_switch.isra.0+0x284/0x9e0 + ? lockdep_hardirqs_on_prepare+0x284/0x400 + ? kthread_is_per_cpu+0xb0/0xb0 + ret_from_fork+0x2d/0x70 + ? kthread_is_per_cpu+0xb0/0xb0 + ret_from_fork_asm+0x11/0x20 + +Fixes: d3d930411ce3 ("RDMA/mlx5: Fix implicit ODP use after free") +Link: https://patch.msgid.link/r/a85ddd16f45c8cb2bc0a188c2b0fcedfce975eb8.1750061791.git.leon@kernel.org +Signed-off-by: Or Har-Toov +Reviewed-by: Patrisious Haddad +Signed-off-by: Leon Romanovsky +Signed-off-by: Jason Gunthorpe +Signed-off-by: Sasha Levin +--- + drivers/infiniband/hw/mlx5/odp.c | 8 ++++---- + 1 file changed, 4 insertions(+), 4 deletions(-) + +diff --git a/drivers/infiniband/hw/mlx5/odp.c b/drivers/infiniband/hw/mlx5/odp.c +index 86d8fa63bf691..4183bab753a39 100644 +--- a/drivers/infiniband/hw/mlx5/odp.c ++++ b/drivers/infiniband/hw/mlx5/odp.c +@@ -247,8 +247,8 @@ static void destroy_unused_implicit_child_mr(struct mlx5_ib_mr *mr) + } + + if (MLX5_CAP_ODP(mr_to_mdev(mr)->mdev, mem_page_fault)) +- __xa_erase(&mr_to_mdev(mr)->odp_mkeys, +- mlx5_base_mkey(mr->mmkey.key)); ++ xa_erase(&mr_to_mdev(mr)->odp_mkeys, ++ mlx5_base_mkey(mr->mmkey.key)); + xa_unlock(&imr->implicit_children); + + /* Freeing a MR is a sleeping operation, so bounce to a work queue */ +@@ -521,8 +521,8 @@ static struct mlx5_ib_mr *implicit_get_child_mr(struct mlx5_ib_mr *imr, + } + + if (MLX5_CAP_ODP(dev->mdev, mem_page_fault)) { +- ret = __xa_store(&dev->odp_mkeys, mlx5_base_mkey(mr->mmkey.key), +- &mr->mmkey, GFP_KERNEL); ++ ret = xa_store(&dev->odp_mkeys, mlx5_base_mkey(mr->mmkey.key), ++ &mr->mmkey, GFP_KERNEL); + if (xa_is_err(ret)) { + ret = ERR_PTR(xa_err(ret)); + __xa_erase(&imr->implicit_children, idx); +-- +2.39.5 + diff --git a/queue-6.15/rdma-mlx5-fix-vport-loopback-for-mpv-device.patch b/queue-6.15/rdma-mlx5-fix-vport-loopback-for-mpv-device.patch new file mode 100644 index 0000000000..0e4d5b95c8 --- /dev/null +++ b/queue-6.15/rdma-mlx5-fix-vport-loopback-for-mpv-device.patch @@ -0,0 +1,87 @@ +From 841df7d1b83ee4c0e14805cab20e2e8d3295a9a2 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Mon, 16 Jun 2025 12:14:54 +0300 +Subject: RDMA/mlx5: Fix vport loopback for MPV device + +From: Patrisious Haddad + +[ Upstream commit a9a9e68954f29b1e197663f76289db4879fd51bb ] + +Always enable vport loopback for both MPV devices on driver start. + +Previously in some cases related to MPV RoCE, packets weren't correctly +executing loopback check at vport in FW, since it was disabled. +Due to complexity of identifying such cases for MPV always enable vport +loopback for both GVMIs when binding the slave to the master port. + +Fixes: 0042f9e458a5 ("RDMA/mlx5: Enable vport loopback when user context or QP mandate") +Signed-off-by: Patrisious Haddad +Reviewed-by: Mark Bloch +Link: https://patch.msgid.link/d4298f5ebb2197459e9e7221c51ecd6a34699847.1750064969.git.leon@kernel.org +Signed-off-by: Leon Romanovsky +Signed-off-by: Sasha Levin +--- + drivers/infiniband/hw/mlx5/main.c | 33 +++++++++++++++++++++++++++++++ + 1 file changed, 33 insertions(+) + +diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c +index d07cacaa0abd0..4ffe3afb560c9 100644 +--- a/drivers/infiniband/hw/mlx5/main.c ++++ b/drivers/infiniband/hw/mlx5/main.c +@@ -1779,6 +1779,33 @@ static void deallocate_uars(struct mlx5_ib_dev *dev, + context->devx_uid); + } + ++static int mlx5_ib_enable_lb_mp(struct mlx5_core_dev *master, ++ struct mlx5_core_dev *slave) ++{ ++ int err; ++ ++ err = mlx5_nic_vport_update_local_lb(master, true); ++ if (err) ++ return err; ++ ++ err = mlx5_nic_vport_update_local_lb(slave, true); ++ if (err) ++ goto out; ++ ++ return 0; ++ ++out: ++ mlx5_nic_vport_update_local_lb(master, false); ++ return err; ++} ++ ++static void mlx5_ib_disable_lb_mp(struct mlx5_core_dev *master, ++ struct mlx5_core_dev *slave) ++{ ++ mlx5_nic_vport_update_local_lb(slave, false); ++ mlx5_nic_vport_update_local_lb(master, false); ++} ++ + int mlx5_ib_enable_lb(struct mlx5_ib_dev *dev, bool td, bool qp) + { + int err = 0; +@@ -3483,6 +3510,8 @@ static void mlx5_ib_unbind_slave_port(struct mlx5_ib_dev *ibdev, + + lockdep_assert_held(&mlx5_ib_multiport_mutex); + ++ mlx5_ib_disable_lb_mp(ibdev->mdev, mpi->mdev); ++ + mlx5_core_mp_event_replay(ibdev->mdev, + MLX5_DRIVER_EVENT_AFFILIATION_REMOVED, + NULL); +@@ -3578,6 +3607,10 @@ static bool mlx5_ib_bind_slave_port(struct mlx5_ib_dev *ibdev, + MLX5_DRIVER_EVENT_AFFILIATION_DONE, + &key); + ++ err = mlx5_ib_enable_lb_mp(ibdev->mdev, mpi->mdev); ++ if (err) ++ goto unbind; ++ + return true; + + unbind: +-- +2.39.5 + diff --git a/queue-6.15/rdma-mlx5-initialize-obj_event-obj_sub_list-before-x.patch b/queue-6.15/rdma-mlx5-initialize-obj_event-obj_sub_list-before-x.patch new file mode 100644 index 0000000000..064f294274 --- /dev/null +++ b/queue-6.15/rdma-mlx5-initialize-obj_event-obj_sub_list-before-x.patch @@ -0,0 +1,100 @@ +From df41eb67cd49e55ec344be5bc740df8799aeedbf Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Tue, 17 Jun 2025 11:13:55 +0300 +Subject: RDMA/mlx5: Initialize obj_event->obj_sub_list before xa_insert + +From: Mark Zhang + +[ Upstream commit 8edab8a72d67742f87e9dc2e2b0cdfddda5dc29a ] + +The obj_event may be loaded immediately after inserted, then if the +list_head is not initialized then we may get a poisonous pointer. This +fixes the crash below: + + mlx5_core 0000:03:00.0: MLX5E: StrdRq(1) RqSz(8) StrdSz(2048) RxCqeCmprss(0 enhanced) + mlx5_core.sf mlx5_core.sf.4: firmware version: 32.38.3056 + mlx5_core 0000:03:00.0 en3f0pf0sf2002: renamed from eth0 + mlx5_core.sf mlx5_core.sf.4: Rate limit: 127 rates are supported, range: 0Mbps to 195312Mbps + IPv6: ADDRCONF(NETDEV_CHANGE): en3f0pf0sf2002: link becomes ready + Unable to handle kernel NULL pointer dereference at virtual address 0000000000000060 + Mem abort info: + ESR = 0x96000006 + EC = 0x25: DABT (current EL), IL = 32 bits + SET = 0, FnV = 0 + EA = 0, S1PTW = 0 + Data abort info: + ISV = 0, ISS = 0x00000006 + CM = 0, WnR = 0 + user pgtable: 4k pages, 48-bit VAs, pgdp=00000007760fb000 + [0000000000000060] pgd=000000076f6d7003, p4d=000000076f6d7003, pud=0000000777841003, pmd=0000000000000000 + Internal error: Oops: 96000006 [#1] SMP + Modules linked in: ipmb_host(OE) act_mirred(E) cls_flower(E) sch_ingress(E) mptcp_diag(E) udp_diag(E) raw_diag(E) unix_diag(E) tcp_diag(E) inet_diag(E) binfmt_misc(E) bonding(OE) rdma_ucm(OE) rdma_cm(OE) iw_cm(OE) ib_ipoib(OE) ib_cm(OE) isofs(E) cdrom(E) mst_pciconf(OE) ib_umad(OE) mlx5_ib(OE) ipmb_dev_int(OE) mlx5_core(OE) kpatch_15237886(OEK) mlxdevm(OE) auxiliary(OE) ib_uverbs(OE) ib_core(OE) psample(E) mlxfw(OE) tls(E) sunrpc(E) vfat(E) fat(E) crct10dif_ce(E) ghash_ce(E) sha1_ce(E) sbsa_gwdt(E) virtio_console(E) ext4(E) mbcache(E) jbd2(E) xfs(E) libcrc32c(E) mmc_block(E) virtio_net(E) net_failover(E) failover(E) sha2_ce(E) sha256_arm64(E) nvme(OE) nvme_core(OE) gpio_mlxbf3(OE) mlx_compat(OE) mlxbf_pmc(OE) i2c_mlxbf(OE) sdhci_of_dwcmshc(OE) pinctrl_mlxbf3(OE) mlxbf_pka(OE) gpio_generic(E) i2c_core(E) mmc_core(E) mlxbf_gige(OE) vitesse(E) pwr_mlxbf(OE) mlxbf_tmfifo(OE) micrel(E) mlxbf_bootctl(OE) virtio_ring(E) virtio(E) ipmi_devintf(E) ipmi_msghandler(E) + [last unloaded: mst_pci] + CPU: 11 PID: 20913 Comm: rte-worker-11 Kdump: loaded Tainted: G OE K 5.10.134-13.1.an8.aarch64 #1 + Hardware name: https://www.mellanox.com BlueField-3 SmartNIC Main Card/BlueField-3 SmartNIC Main Card, BIOS 4.2.2.12968 Oct 26 2023 + pstate: a0400089 (NzCv daIf +PAN -UAO -TCO BTYPE=--) + pc : dispatch_event_fd+0x68/0x300 [mlx5_ib] + lr : devx_event_notifier+0xcc/0x228 [mlx5_ib] + sp : ffff80001005bcf0 + x29: ffff80001005bcf0 x28: 0000000000000001 + x27: ffff244e0740a1d8 x26: ffff244e0740a1d0 + x25: ffffda56beff5ae0 x24: ffffda56bf911618 + x23: ffff244e0596a480 x22: ffff244e0596a480 + x21: ffff244d8312ad90 x20: ffff244e0596a480 + x19: fffffffffffffff0 x18: 0000000000000000 + x17: 0000000000000000 x16: ffffda56be66d620 + x15: 0000000000000000 x14: 0000000000000000 + x13: 0000000000000000 x12: 0000000000000000 + x11: 0000000000000040 x10: ffffda56bfcafb50 + x9 : ffffda5655c25f2c x8 : 0000000000000010 + x7 : 0000000000000000 x6 : ffff24545a2e24b8 + x5 : 0000000000000003 x4 : ffff80001005bd28 + x3 : 0000000000000000 x2 : 0000000000000000 + x1 : ffff244e0596a480 x0 : ffff244d8312ad90 + Call trace: + dispatch_event_fd+0x68/0x300 [mlx5_ib] + devx_event_notifier+0xcc/0x228 [mlx5_ib] + atomic_notifier_call_chain+0x58/0x80 + mlx5_eq_async_int+0x148/0x2b0 [mlx5_core] + atomic_notifier_call_chain+0x58/0x80 + irq_int_handler+0x20/0x30 [mlx5_core] + __handle_irq_event_percpu+0x60/0x220 + handle_irq_event_percpu+0x3c/0x90 + handle_irq_event+0x58/0x158 + handle_fasteoi_irq+0xfc/0x188 + generic_handle_irq+0x34/0x48 + ... + +Fixes: 759738537142 ("IB/mlx5: Enable subscription for device events over DEVX") +Link: https://patch.msgid.link/r/3ce7f20e0d1a03dc7de6e57494ec4b8eaf1f05c2.1750147949.git.leon@kernel.org +Signed-off-by: Mark Zhang +Signed-off-by: Leon Romanovsky +Signed-off-by: Jason Gunthorpe +Signed-off-by: Sasha Levin +--- + drivers/infiniband/hw/mlx5/devx.c | 2 +- + 1 file changed, 1 insertion(+), 1 deletion(-) + +diff --git a/drivers/infiniband/hw/mlx5/devx.c b/drivers/infiniband/hw/mlx5/devx.c +index bceae1c1f9801..843dcd3122424 100644 +--- a/drivers/infiniband/hw/mlx5/devx.c ++++ b/drivers/infiniband/hw/mlx5/devx.c +@@ -1958,6 +1958,7 @@ subscribe_event_xa_alloc(struct mlx5_devx_event_table *devx_event_table, + /* Level1 is valid for future use, no need to free */ + return -ENOMEM; + ++ INIT_LIST_HEAD(&obj_event->obj_sub_list); + err = xa_insert(&event->object_ids, + key_level2, + obj_event, +@@ -1966,7 +1967,6 @@ subscribe_event_xa_alloc(struct mlx5_devx_event_table *devx_event_table, + kfree(obj_event); + return err; + } +- INIT_LIST_HEAD(&obj_event->obj_sub_list); + } + + return 0; +-- +2.39.5 + diff --git a/queue-6.15/rdma-mlx5-reduce-stack-usage-in-mlx5_ib_ufile_hw_cle.patch b/queue-6.15/rdma-mlx5-reduce-stack-usage-in-mlx5_ib_ufile_hw_cle.patch new file mode 100644 index 0000000000..4cb666bb3a --- /dev/null +++ b/queue-6.15/rdma-mlx5-reduce-stack-usage-in-mlx5_ib_ufile_hw_cle.patch @@ -0,0 +1,65 @@ +From 9b051f105741f5e2b93fc555116400487585f0bc Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Tue, 10 Jun 2025 11:28:42 +0200 +Subject: RDMA/mlx5: reduce stack usage in mlx5_ib_ufile_hw_cleanup + +From: Arnd Bergmann + +[ Upstream commit b26852daaa83f535109253d114426d1fa674155d ] + +This function has an array of eight mlx5_async_cmd structures, which +often fits on the stack, but depending on the configuration can +end up blowing the stack frame warning limit: + +drivers/infiniband/hw/mlx5/devx.c:2670:6: error: stack frame size (1392) exceeds limit (1280) in 'mlx5_ib_ufile_hw_cleanup' [-Werror,-Wframe-larger-than] + +Change this to a dynamic allocation instead. While a kmalloc() +can theoretically fail, a GFP_KERNEL allocation under a page will +block until memory has been freed up, so in the worst case, this +only adds extra time in an already constrained environment. + +Fixes: 7c891a4dbcc1 ("RDMA/mlx5: Add implementation for ufile_hw_cleanup device operation") +Signed-off-by: Arnd Bergmann +Link: https://patch.msgid.link/20250610092846.2642535-1-arnd@kernel.org +Signed-off-by: Leon Romanovsky +Signed-off-by: Sasha Levin +--- + drivers/infiniband/hw/mlx5/devx.c | 8 +++++++- + 1 file changed, 7 insertions(+), 1 deletion(-) + +diff --git a/drivers/infiniband/hw/mlx5/devx.c b/drivers/infiniband/hw/mlx5/devx.c +index 2479da8620ca9..bceae1c1f9801 100644 +--- a/drivers/infiniband/hw/mlx5/devx.c ++++ b/drivers/infiniband/hw/mlx5/devx.c +@@ -2669,7 +2669,7 @@ static void devx_wait_async_destroy(struct mlx5_async_cmd *cmd) + + void mlx5_ib_ufile_hw_cleanup(struct ib_uverbs_file *ufile) + { +- struct mlx5_async_cmd async_cmd[MAX_ASYNC_CMDS]; ++ struct mlx5_async_cmd *async_cmd; + struct ib_ucontext *ucontext = ufile->ucontext; + struct ib_device *device = ucontext->device; + struct mlx5_ib_dev *dev = to_mdev(device); +@@ -2678,6 +2678,10 @@ void mlx5_ib_ufile_hw_cleanup(struct ib_uverbs_file *ufile) + int head = 0; + int tail = 0; + ++ async_cmd = kcalloc(MAX_ASYNC_CMDS, sizeof(*async_cmd), GFP_KERNEL); ++ if (!async_cmd) ++ return; ++ + list_for_each_entry(uobject, &ufile->uobjects, list) { + WARN_ON(uverbs_try_lock_object(uobject, UVERBS_LOOKUP_WRITE)); + +@@ -2713,6 +2717,8 @@ void mlx5_ib_ufile_hw_cleanup(struct ib_uverbs_file *ufile) + devx_wait_async_destroy(&async_cmd[head % MAX_ASYNC_CMDS]); + head++; + } ++ ++ kfree(async_cmd); + } + + static ssize_t devx_async_cmd_event_read(struct file *filp, char __user *buf, +-- +2.39.5 + diff --git a/queue-6.15/regulator-fan53555-add-enable_time-support-and-soft-.patch b/queue-6.15/regulator-fan53555-add-enable_time-support-and-soft-.patch new file mode 100644 index 0000000000..6f527f9a0f --- /dev/null +++ b/queue-6.15/regulator-fan53555-add-enable_time-support-and-soft-.patch @@ -0,0 +1,149 @@ +From bb65f4f470327fbbd758947257329973777048e6 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Fri, 6 Jun 2025 21:04:18 +0200 +Subject: regulator: fan53555: add enable_time support and soft-start times + +From: Heiko Stuebner + +[ Upstream commit 8acfb165a492251a08a22a4fa6497a131e8c2609 ] + +The datasheets for all the fan53555 variants (and clones using the same +interface) define so called soft start times, from enabling the regulator +until at least some percentage of the output (i.e. 92% for the rk860x +types) are available. + +The regulator framework supports this with the enable_time property +but currently the fan53555 driver does not define enable_times for any +variant. + +I ran into a problem with this while testing the new driver for the +Rockchip NPUs (rocket), which does runtime-pm including disabling and +enabling a rk8602 as needed. When reenabling the regulator while running +a load, fatal hangs could be observed while enabling the associated +power-domain, which the regulator supplies. + +Experimentally setting the regulator to always-on, made the issue +disappear, leading to the missing delay to let power stabilize. +And as expected, setting the enable-time to a non-zero value +according to the datasheet also resolved the regulator-issue. + +The datasheets in nearly all cases only specify "typical" values, +except for the fan53555 type 08. There both a typical and maximum +value are listed - 40uS apart. + +For all typical values I've added 100uS to be on the safe side. +Individual details for the relevant regulators below: + +- fan53526: + The datasheet for all variants lists a typical value of 150uS, so + make that 250uS with safety margin. +- fan53555: + types 08 and 18 (unsupported) are given a typical enable time of 135uS + but also a maximum of 175uS so use that value. All the other types only + have a typical time in the datasheet of 300uS, so give a bit margin by + setting it to 400uS. +- rk8600 + rk8602: + Datasheet reports a typical value of 260us, so use 360uS to be safe. +- syr82x + syr83x: + All datasheets report typical soft-start values of 300uS for these + regulators, so use 400uS. +- tcs452x: + Datasheet sadly does not report a soft-start time, so I've not set + an enable-time + +Signed-off-by: Heiko Stuebner +Link: https://patch.msgid.link/20250606190418.478633-1-heiko@sntech.de +Signed-off-by: Mark Brown +Signed-off-by: Sasha Levin +--- + drivers/regulator/fan53555.c | 14 ++++++++++++++ + 1 file changed, 14 insertions(+) + +diff --git a/drivers/regulator/fan53555.c b/drivers/regulator/fan53555.c +index bd9447dac5967..c282236959b18 100644 +--- a/drivers/regulator/fan53555.c ++++ b/drivers/regulator/fan53555.c +@@ -147,6 +147,7 @@ struct fan53555_device_info { + unsigned int slew_mask; + const unsigned int *ramp_delay_table; + unsigned int n_ramp_values; ++ unsigned int enable_time; + unsigned int slew_rate; + }; + +@@ -282,6 +283,7 @@ static int fan53526_voltages_setup_fairchild(struct fan53555_device_info *di) + di->slew_mask = CTL_SLEW_MASK; + di->ramp_delay_table = slew_rates; + di->n_ramp_values = ARRAY_SIZE(slew_rates); ++ di->enable_time = 250; + di->vsel_count = FAN53526_NVOLTAGES; + + return 0; +@@ -296,10 +298,12 @@ static int fan53555_voltages_setup_fairchild(struct fan53555_device_info *di) + case FAN53555_CHIP_REV_00: + di->vsel_min = 600000; + di->vsel_step = 10000; ++ di->enable_time = 400; + break; + case FAN53555_CHIP_REV_13: + di->vsel_min = 800000; + di->vsel_step = 10000; ++ di->enable_time = 400; + break; + default: + dev_err(di->dev, +@@ -311,13 +315,19 @@ static int fan53555_voltages_setup_fairchild(struct fan53555_device_info *di) + case FAN53555_CHIP_ID_01: + case FAN53555_CHIP_ID_03: + case FAN53555_CHIP_ID_05: ++ di->vsel_min = 600000; ++ di->vsel_step = 10000; ++ di->enable_time = 400; ++ break; + case FAN53555_CHIP_ID_08: + di->vsel_min = 600000; + di->vsel_step = 10000; ++ di->enable_time = 175; + break; + case FAN53555_CHIP_ID_04: + di->vsel_min = 603000; + di->vsel_step = 12826; ++ di->enable_time = 400; + break; + default: + dev_err(di->dev, +@@ -350,6 +360,7 @@ static int fan53555_voltages_setup_rockchip(struct fan53555_device_info *di) + di->slew_mask = CTL_SLEW_MASK; + di->ramp_delay_table = slew_rates; + di->n_ramp_values = ARRAY_SIZE(slew_rates); ++ di->enable_time = 360; + di->vsel_count = FAN53555_NVOLTAGES; + + return 0; +@@ -372,6 +383,7 @@ static int rk8602_voltages_setup_rockchip(struct fan53555_device_info *di) + di->slew_mask = CTL_SLEW_MASK; + di->ramp_delay_table = slew_rates; + di->n_ramp_values = ARRAY_SIZE(slew_rates); ++ di->enable_time = 360; + di->vsel_count = RK8602_NVOLTAGES; + + return 0; +@@ -395,6 +407,7 @@ static int fan53555_voltages_setup_silergy(struct fan53555_device_info *di) + di->slew_mask = CTL_SLEW_MASK; + di->ramp_delay_table = slew_rates; + di->n_ramp_values = ARRAY_SIZE(slew_rates); ++ di->enable_time = 400; + di->vsel_count = FAN53555_NVOLTAGES; + + return 0; +@@ -594,6 +607,7 @@ static int fan53555_regulator_register(struct fan53555_device_info *di, + rdesc->ramp_mask = di->slew_mask; + rdesc->ramp_delay_table = di->ramp_delay_table; + rdesc->n_ramp_values = di->n_ramp_values; ++ rdesc->enable_time = di->enable_time; + rdesc->owner = THIS_MODULE; + + rdev = devm_regulator_register(di->dev, &di->desc, config); +-- +2.39.5 + diff --git a/queue-6.15/rose-fix-dangling-neighbour-pointers-in-rose_rt_devi.patch b/queue-6.15/rose-fix-dangling-neighbour-pointers-in-rose_rt_devi.patch new file mode 100644 index 0000000000..d849b22125 --- /dev/null +++ b/queue-6.15/rose-fix-dangling-neighbour-pointers-in-rose_rt_devi.patch @@ -0,0 +1,84 @@ +From bde5eeed317a4249dff5ce8a4b947b5f60204656 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Sun, 29 Jun 2025 12:06:31 +0900 +Subject: rose: fix dangling neighbour pointers in rose_rt_device_down() + +From: Kohei Enju + +[ Upstream commit 34a500caf48c47d5171f4aa1f237da39b07c6157 ] + +There are two bugs in rose_rt_device_down() that can cause +use-after-free: + +1. The loop bound `t->count` is modified within the loop, which can + cause the loop to terminate early and miss some entries. + +2. When removing an entry from the neighbour array, the subsequent entries + are moved up to fill the gap, but the loop index `i` is still + incremented, causing the next entry to be skipped. + +For example, if a node has three neighbours (A, A, B) with count=3 and A +is being removed, the second A is not checked. + + i=0: (A, A, B) -> (A, B) with count=2 + ^ checked + i=1: (A, B) -> (A, B) with count=2 + ^ checked (B, not A!) + i=2: (doesn't occur because i < count is false) + +This leaves the second A in the array with count=2, but the rose_neigh +structure has been freed. Code that accesses these entries assumes that +the first `count` entries are valid pointers, causing a use-after-free +when it accesses the dangling pointer. + +Fix both issues by iterating over the array in reverse order with a fixed +loop bound. This ensures that all entries are examined and that the removal +of an entry doesn't affect subsequent iterations. + +Reported-by: syzbot+e04e2c007ba2c80476cb@syzkaller.appspotmail.com +Closes: https://syzkaller.appspot.com/bug?extid=e04e2c007ba2c80476cb +Tested-by: syzbot+e04e2c007ba2c80476cb@syzkaller.appspotmail.com +Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") +Signed-off-by: Kohei Enju +Reviewed-by: Simon Horman +Link: https://patch.msgid.link/20250629030833.6680-1-enjuk@amazon.com +Signed-off-by: Jakub Kicinski +Signed-off-by: Sasha Levin +--- + net/rose/rose_route.c | 15 ++++----------- + 1 file changed, 4 insertions(+), 11 deletions(-) + +diff --git a/net/rose/rose_route.c b/net/rose/rose_route.c +index 2dd6bd3a3011f..b72bf8a08d489 100644 +--- a/net/rose/rose_route.c ++++ b/net/rose/rose_route.c +@@ -497,22 +497,15 @@ void rose_rt_device_down(struct net_device *dev) + t = rose_node; + rose_node = rose_node->next; + +- for (i = 0; i < t->count; i++) { ++ for (i = t->count - 1; i >= 0; i--) { + if (t->neighbour[i] != s) + continue; + + t->count--; + +- switch (i) { +- case 0: +- t->neighbour[0] = t->neighbour[1]; +- fallthrough; +- case 1: +- t->neighbour[1] = t->neighbour[2]; +- break; +- case 2: +- break; +- } ++ memmove(&t->neighbour[i], &t->neighbour[i + 1], ++ sizeof(t->neighbour[0]) * ++ (t->count - i)); + } + + if (t->count <= 0) +-- +2.39.5 + diff --git a/queue-6.15/scsi-core-enforce-unlimited-max_segment_size-when-vi.patch b/queue-6.15/scsi-core-enforce-unlimited-max_segment_size-when-vi.patch new file mode 100644 index 0000000000..34b40f19fc --- /dev/null +++ b/queue-6.15/scsi-core-enforce-unlimited-max_segment_size-when-vi.patch @@ -0,0 +1,83 @@ +From c317f60ff1e8151ed1431177d05d5ed5eb3fbcfa Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Tue, 24 Jun 2025 14:52:28 +0200 +Subject: scsi: core: Enforce unlimited max_segment_size when + virt_boundary_mask is set + +From: Christoph Hellwig + +[ Upstream commit 4937e604ca24c41cae3296d069c871c2f3f519c8 ] + +The virt_boundary_mask limit requires an unlimited max_segment_size for +bio splitting to not corrupt data. Historically, the block layer tried +to validate this, although the check was half-hearted until the addition +of the atomic queue limits API. The full blown check then triggered +issues with stacked devices incorrectly inheriting limits such as the +virt boundary and got disabled in commit b561ea56a264 ("block: allow +device to have both virt_boundary_mask and max segment size") instead of +fixing the issue properly. + +Ensure that the SCSI mid layer doesn't set the default low +max_segment_size limit for this case, and check for invalid +max_segment_size values in the host template, similar to the original +block layer check given that SCSI devices can't be stacked. + +This fixes reported data corruption on storvsc, although as far as I can +tell storvsc always failed to properly set the max_segment_size limit as +the SCSI APIs historically applied that when setting up the host, while +storvsc only set the virt_boundary_mask when configuring the scsi_device. + +Fixes: 81988a0e6b03 ("storvsc: get rid of bounce buffer") +Fixes: b561ea56a264 ("block: allow device to have both virt_boundary_mask and max segment size") +Reported-by: Ming Lei +Signed-off-by: Christoph Hellwig +Link: https://lore.kernel.org/r/20250624125233.219635-3-hch@lst.de +Reviewed-by: John Garry +Reviewed-by: Ming Lei +Reviewed-by: Hannes Reinecke +Reviewed-by: Bart Van Assche +Signed-off-by: Martin K. Petersen +Signed-off-by: Sasha Levin +--- + drivers/scsi/hosts.c | 18 +++++++++++------- + 1 file changed, 11 insertions(+), 7 deletions(-) + +diff --git a/drivers/scsi/hosts.c b/drivers/scsi/hosts.c +index e021f1106beab..cc5d05dc395c4 100644 +--- a/drivers/scsi/hosts.c ++++ b/drivers/scsi/hosts.c +@@ -473,10 +473,17 @@ struct Scsi_Host *scsi_host_alloc(const struct scsi_host_template *sht, int priv + else + shost->max_sectors = SCSI_DEFAULT_MAX_SECTORS; + +- if (sht->max_segment_size) +- shost->max_segment_size = sht->max_segment_size; +- else +- shost->max_segment_size = BLK_MAX_SEGMENT_SIZE; ++ shost->virt_boundary_mask = sht->virt_boundary_mask; ++ if (shost->virt_boundary_mask) { ++ WARN_ON_ONCE(sht->max_segment_size && ++ sht->max_segment_size != UINT_MAX); ++ shost->max_segment_size = UINT_MAX; ++ } else { ++ if (sht->max_segment_size) ++ shost->max_segment_size = sht->max_segment_size; ++ else ++ shost->max_segment_size = BLK_MAX_SEGMENT_SIZE; ++ } + + /* 32-byte (dword) is a common minimum for HBAs. */ + if (sht->dma_alignment) +@@ -492,9 +499,6 @@ struct Scsi_Host *scsi_host_alloc(const struct scsi_host_template *sht, int priv + else + shost->dma_boundary = 0xffffffff; + +- if (sht->virt_boundary_mask) +- shost->virt_boundary_mask = sht->virt_boundary_mask; +- + device_initialize(&shost->shost_gendev); + dev_set_name(&shost->shost_gendev, "host%d", shost->host_no); + shost->shost_gendev.bus = &scsi_bus_type; +-- +2.39.5 + diff --git a/queue-6.15/scsi-qla2xxx-fix-dma-mapping-test-in-qla24xx_get_por.patch b/queue-6.15/scsi-qla2xxx-fix-dma-mapping-test-in-qla24xx_get_por.patch new file mode 100644 index 0000000000..32261dd636 --- /dev/null +++ b/queue-6.15/scsi-qla2xxx-fix-dma-mapping-test-in-qla24xx_get_por.patch @@ -0,0 +1,38 @@ +From 42a157fed15df280a15bd1cd1f771977828d9c3d Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Tue, 17 Jun 2025 18:11:11 +0200 +Subject: scsi: qla2xxx: Fix DMA mapping test in qla24xx_get_port_database() + +From: Thomas Fourier + +[ Upstream commit c3b214719a87735d4f67333a8ef3c0e31a34837c ] + +dma_map_XXX() functions return as error values DMA_MAPPING_ERROR which is +often ~0. The error value should be tested with dma_mapping_error() like +it was done in qla26xx_dport_diagnostics(). + +Fixes: 818c7f87a177 ("scsi: qla2xxx: Add changes in preparation for vendor extended FDMI/RDP") +Signed-off-by: Thomas Fourier +Link: https://lore.kernel.org/r/20250617161115.39888-2-fourier.thomas@gmail.com +Signed-off-by: Martin K. Petersen +Signed-off-by: Sasha Levin +--- + drivers/scsi/qla2xxx/qla_mbx.c | 2 +- + 1 file changed, 1 insertion(+), 1 deletion(-) + +diff --git a/drivers/scsi/qla2xxx/qla_mbx.c b/drivers/scsi/qla2xxx/qla_mbx.c +index 0cd6f3e148824..13b6cb1b93acd 100644 +--- a/drivers/scsi/qla2xxx/qla_mbx.c ++++ b/drivers/scsi/qla2xxx/qla_mbx.c +@@ -2147,7 +2147,7 @@ qla24xx_get_port_database(scsi_qla_host_t *vha, u16 nport_handle, + + pdb_dma = dma_map_single(&vha->hw->pdev->dev, pdb, + sizeof(*pdb), DMA_FROM_DEVICE); +- if (!pdb_dma) { ++ if (dma_mapping_error(&vha->hw->pdev->dev, pdb_dma)) { + ql_log(ql_log_warn, vha, 0x1116, "Failed to map dma buffer.\n"); + return QLA_MEMORY_ALLOC_FAILED; + } +-- +2.39.5 + diff --git a/queue-6.15/scsi-qla4xxx-fix-missing-dma-mapping-error-in-qla4xx.patch b/queue-6.15/scsi-qla4xxx-fix-missing-dma-mapping-error-in-qla4xx.patch new file mode 100644 index 0000000000..c8c7057e49 --- /dev/null +++ b/queue-6.15/scsi-qla4xxx-fix-missing-dma-mapping-error-in-qla4xx.patch @@ -0,0 +1,37 @@ +From cb5d7123c50c9f4121fbdf1fda03011e41d267d7 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Wed, 18 Jun 2025 09:17:37 +0200 +Subject: scsi: qla4xxx: Fix missing DMA mapping error in qla4xxx_alloc_pdu() + +From: Thomas Fourier + +[ Upstream commit 00f452a1b084efbe8dcb60a29860527944a002a1 ] + +dma_map_XXX() can fail and should be tested for errors with +dma_mapping_error(). + +Fixes: b3a271a94d00 ("[SCSI] qla4xxx: support iscsiadm session mgmt") +Signed-off-by: Thomas Fourier +Link: https://lore.kernel.org/r/20250618071742.21822-2-fourier.thomas@gmail.com +Signed-off-by: Martin K. Petersen +Signed-off-by: Sasha Levin +--- + drivers/scsi/qla4xxx/ql4_os.c | 2 ++ + 1 file changed, 2 insertions(+) + +diff --git a/drivers/scsi/qla4xxx/ql4_os.c b/drivers/scsi/qla4xxx/ql4_os.c +index d540d66e6ffc4..80c4b4e7526b1 100644 +--- a/drivers/scsi/qla4xxx/ql4_os.c ++++ b/drivers/scsi/qla4xxx/ql4_os.c +@@ -3420,6 +3420,8 @@ static int qla4xxx_alloc_pdu(struct iscsi_task *task, uint8_t opcode) + task_data->data_dma = dma_map_single(&ha->pdev->dev, task->data, + task->data_count, + DMA_TO_DEVICE); ++ if (dma_mapping_error(&ha->pdev->dev, task_data->data_dma)) ++ return -ENOMEM; + } + + DEBUG2(ql4_printk(KERN_INFO, ha, "%s: MaxRecvLen %u, iscsi hrd %d\n", +-- +2.39.5 + diff --git a/queue-6.15/scsi-sd-fix-vpd-page-0xb7-length-check.patch b/queue-6.15/scsi-sd-fix-vpd-page-0xb7-length-check.patch new file mode 100644 index 0000000000..93761d2def --- /dev/null +++ b/queue-6.15/scsi-sd-fix-vpd-page-0xb7-length-check.patch @@ -0,0 +1,47 @@ +From 6ec0a3e1dd82f33fe053ab5593385a0542890c8a Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Thu, 19 Jun 2025 12:03:02 +0800 +Subject: scsi: sd: Fix VPD page 0xb7 length check + +From: jackysliu <1972843537@qq.com> + +[ Upstream commit 8889676cd62161896f1d861ce294adc29c4f2cb5 ] + +sd_read_block_limits_ext() currently assumes that vpd->len excludes the +size of the page header. However, vpd->len describes the size of the entire +VPD page, therefore the sanity check is incorrect. + +In practice this is not really a problem since we don't attach VPD +pages unless they actually report data trailing the header. But fix +the length check regardless. + +This issue was identified by Wukong-Agent (formerly Tencent Woodpecker), a +code security AI agent, through static code analysis. + +[mkp: rewrote patch description] + +Signed-off-by: jackysliu <1972843537@qq.com> +Link: https://lore.kernel.org/r/tencent_ADA5210D1317EEB6CD7F3DE9FE9DA4591D05@qq.com +Fixes: 96b171d6dba6 ("scsi: core: Query the Block Limits Extension VPD page") +Signed-off-by: Martin K. Petersen +Signed-off-by: Sasha Levin +--- + drivers/scsi/sd.c | 2 +- + 1 file changed, 1 insertion(+), 1 deletion(-) + +diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c +index 950d8c9fb8843..89d5c4b17bc46 100644 +--- a/drivers/scsi/sd.c ++++ b/drivers/scsi/sd.c +@@ -3384,7 +3384,7 @@ static void sd_read_block_limits_ext(struct scsi_disk *sdkp) + + rcu_read_lock(); + vpd = rcu_dereference(sdkp->device->vpd_pgb7); +- if (vpd && vpd->len >= 2) ++ if (vpd && vpd->len >= 6) + sdkp->rscs = vpd->data[5] & 1; + rcu_read_unlock(); + } +-- +2.39.5 + diff --git a/queue-6.15/scsi-target-fix-null-pointer-dereference-in-core_scs.patch b/queue-6.15/scsi-target-fix-null-pointer-dereference-in-core_scs.patch new file mode 100644 index 0000000000..6925251d81 --- /dev/null +++ b/queue-6.15/scsi-target-fix-null-pointer-dereference-in-core_scs.patch @@ -0,0 +1,56 @@ +From 9765587c6fc7c5795fa7c317d189b5bc41b2e65d Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Thu, 12 Jun 2025 12:15:56 +0200 +Subject: scsi: target: Fix NULL pointer dereference in + core_scsi3_decode_spec_i_port() + +From: Maurizio Lombardi + +[ Upstream commit d8ab68bdb294b09a761e967dad374f2965e1913f ] + +The function core_scsi3_decode_spec_i_port(), in its error code path, +unconditionally calls core_scsi3_lunacl_undepend_item() passing the +dest_se_deve pointer, which may be NULL. + +This can lead to a NULL pointer dereference if dest_se_deve remains +unset. + +SPC-3 PR SPEC_I_PT: Unable to locate dest_tpg +Unable to handle kernel paging request at virtual address dfff800000000012 +Call trace: + core_scsi3_lunacl_undepend_item+0x2c/0xf0 [target_core_mod] (P) + core_scsi3_decode_spec_i_port+0x120c/0x1c30 [target_core_mod] + core_scsi3_emulate_pro_register+0x6b8/0xcd8 [target_core_mod] + target_scsi3_emulate_pr_out+0x56c/0x840 [target_core_mod] + +Fix this by adding a NULL check before calling +core_scsi3_lunacl_undepend_item() + +Signed-off-by: Maurizio Lombardi +Link: https://lore.kernel.org/r/20250612101556.24829-1-mlombard@redhat.com +Reviewed-by: Mike Christie +Reviewed-by: John Meneghini +Signed-off-by: Martin K. Petersen +Signed-off-by: Sasha Levin +--- + drivers/target/target_core_pr.c | 4 +++- + 1 file changed, 3 insertions(+), 1 deletion(-) + +diff --git a/drivers/target/target_core_pr.c b/drivers/target/target_core_pr.c +index 34cf2c399b399..70905805cb175 100644 +--- a/drivers/target/target_core_pr.c ++++ b/drivers/target/target_core_pr.c +@@ -1842,7 +1842,9 @@ core_scsi3_decode_spec_i_port( + } + + kmem_cache_free(t10_pr_reg_cache, dest_pr_reg); +- core_scsi3_lunacl_undepend_item(dest_se_deve); ++ ++ if (dest_se_deve) ++ core_scsi3_lunacl_undepend_item(dest_se_deve); + + if (is_local) + continue; +-- +2.39.5 + diff --git a/queue-6.15/scsi-ufs-core-fix-spelling-of-a-sysfs-attribute-name.patch b/queue-6.15/scsi-ufs-core-fix-spelling-of-a-sysfs-attribute-name.patch new file mode 100644 index 0000000000..d471b3f550 --- /dev/null +++ b/queue-6.15/scsi-ufs-core-fix-spelling-of-a-sysfs-attribute-name.patch @@ -0,0 +1,60 @@ +From 2f064207c5d462966591913fbaad9e4cef5bf337 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Tue, 24 Jun 2025 11:16:44 -0700 +Subject: scsi: ufs: core: Fix spelling of a sysfs attribute name + +From: Bart Van Assche + +[ Upstream commit 021f243627ead17eb6500170256d3d9be787dad8 ] + +Change "resourse" into "resource" in the name of a sysfs attribute. + +Fixes: d829fc8a1058 ("scsi: ufs: sysfs: unit descriptor") +Signed-off-by: Bart Van Assche +Link: https://lore.kernel.org/r/20250624181658.336035-1-bvanassche@acm.org +Reviewed-by: Avri Altman +Signed-off-by: Martin K. Petersen +Signed-off-by: Sasha Levin +--- + Documentation/ABI/testing/sysfs-driver-ufs | 2 +- + drivers/ufs/core/ufs-sysfs.c | 4 ++-- + 2 files changed, 3 insertions(+), 3 deletions(-) + +diff --git a/Documentation/ABI/testing/sysfs-driver-ufs b/Documentation/ABI/testing/sysfs-driver-ufs +index e36d2de16cbda..0397664e869a0 100644 +--- a/Documentation/ABI/testing/sysfs-driver-ufs ++++ b/Documentation/ABI/testing/sysfs-driver-ufs +@@ -711,7 +711,7 @@ Description: This file shows the thin provisioning type. This is one of + + The file is read only. + +-What: /sys/class/scsi_device/*/device/unit_descriptor/physical_memory_resourse_count ++What: /sys/class/scsi_device/*/device/unit_descriptor/physical_memory_resource_count + Date: February 2018 + Contact: Stanislav Nijnikov + Description: This file shows the total physical memory resources. This is +diff --git a/drivers/ufs/core/ufs-sysfs.c b/drivers/ufs/core/ufs-sysfs.c +index 634cf163f4cb1..c53f924f7fa83 100644 +--- a/drivers/ufs/core/ufs-sysfs.c ++++ b/drivers/ufs/core/ufs-sysfs.c +@@ -1675,7 +1675,7 @@ UFS_UNIT_DESC_PARAM(logical_block_size, _LOGICAL_BLK_SIZE, 1); + UFS_UNIT_DESC_PARAM(logical_block_count, _LOGICAL_BLK_COUNT, 8); + UFS_UNIT_DESC_PARAM(erase_block_size, _ERASE_BLK_SIZE, 4); + UFS_UNIT_DESC_PARAM(provisioning_type, _PROVISIONING_TYPE, 1); +-UFS_UNIT_DESC_PARAM(physical_memory_resourse_count, _PHY_MEM_RSRC_CNT, 8); ++UFS_UNIT_DESC_PARAM(physical_memory_resource_count, _PHY_MEM_RSRC_CNT, 8); + UFS_UNIT_DESC_PARAM(context_capabilities, _CTX_CAPABILITIES, 2); + UFS_UNIT_DESC_PARAM(large_unit_granularity, _LARGE_UNIT_SIZE_M1, 1); + UFS_UNIT_DESC_PARAM(wb_buf_alloc_units, _WB_BUF_ALLOC_UNITS, 4); +@@ -1692,7 +1692,7 @@ static struct attribute *ufs_sysfs_unit_descriptor[] = { + &dev_attr_logical_block_count.attr, + &dev_attr_erase_block_size.attr, + &dev_attr_provisioning_type.attr, +- &dev_attr_physical_memory_resourse_count.attr, ++ &dev_attr_physical_memory_resource_count.attr, + &dev_attr_context_capabilities.attr, + &dev_attr_large_unit_granularity.attr, + &dev_attr_wb_buf_alloc_units.attr, +-- +2.39.5 + diff --git a/queue-6.15/series b/queue-6.15/series index c5dcff0bd2..c974e338a5 100644 --- a/queue-6.15/series +++ b/queue-6.15/series @@ -30,3 +30,111 @@ mtk-sd-fix-a-pagefault-in-dma_unmap_sg-for-not-prepared-data.patch mtk-sd-prevent-memory-corruption-from-dma-map-failure.patch mtk-sd-reset-host-mrq-on-prepare_data-error.patch drm-v3d-disable-interrupts-before-resetting-the-gpu.patch +firmware-arm_ffa-fix-memory-leak-by-freeing-notifier.patch +firmware-arm_ffa-move-memory-allocation-outside-the-.patch +firmware-arm_ffa-replace-mutex-with-rwlock-to-avoid-.patch +firmware-arm_ffa-fix-the-missing-entry-in-struct-ffa.patch +firmware-exynos-acpm-fix-timeouts-on-xfers-handling.patch +rdma-mlx5-reduce-stack-usage-in-mlx5_ib_ufile_hw_cle.patch +arm64-dts-apple-t8103-fix-pcie-bcm4377-nodename.patch +arm64-dts-apple-drop-address-size-cells-from-spi-nor.patch +arm64-dts-apple-move-touchbar-mipi-address-size-cell.patch +platform-mellanox-mlxbf-tmfifo-fix-vring_desc.len-as.patch +rdma-mlx5-fix-unsafe-xarray-access-in-implicit-odp-h.patch +rdma-mlx5-initialize-obj_event-obj_sub_list-before-x.patch +module-provide-export_symbol_gpl_for_modules-helper.patch +fs-export-anon_inode_make_secure_inode-and-fix-secre.patch +nfs-clean-up-proc-net-rpc-nfs-when-nfs_fs_proc_net_i.patch +nfsv4-pnfs-fix-a-race-to-wake-on-nfs_layout_drain.patch +scsi-qla2xxx-fix-dma-mapping-test-in-qla24xx_get_por.patch +scsi-qla4xxx-fix-missing-dma-mapping-error-in-qla4xx.patch +scsi-sd-fix-vpd-page-0xb7-length-check.patch +scsi-core-enforce-unlimited-max_segment_size-when-vi.patch +scsi-ufs-core-fix-spelling-of-a-sysfs-attribute-name.patch +ib-mlx5-fix-potential-deadlock-in-mr-deregistration.patch +rdma-mlx5-fix-hw-counters-query-for-non-representor-.patch +rdma-mlx5-fix-cc-counters-query-for-mpv.patch +rdma-mlx5-fix-vport-loopback-for-mpv-device.patch +platform-mellanox-mlxbf-pmc-fix-duplicate-event-id-f.patch +platform-mellanox-nvsw-sn2201-fix-bus-number-in-adap.patch +platform-x86-wmi-fix-wmi-event-enablement.patch +flexfiles-pnfs-update-stats-on-nfs4err_delay-for-v4..patch +nfsv4-flexfiles-fix-handling-of-nfs-level-errors-in-.patch +bluetooth-prevent-unintended-pause-by-checking-if-ad.patch +btrfs-fix-failure-to-rebuild-free-space-tree-using-m.patch +btrfs-fix-missing-error-handling-when-searching-for-.patch +btrfs-fix-iteration-of-extrefs-during-log-replay.patch +btrfs-fix-inode-lookup-error-handling-during-log-rep.patch +btrfs-record-new-subvolume-in-parent-dir-earlier-to-.patch +btrfs-propagate-last_unlink_trans-earlier-when-doing.patch +btrfs-use-btrfs_record_snapshot_destroy-during-rmdir.patch +ethernet-atl1-add-missing-dma-mapping-error-checks-a.patch +dpaa2-eth-fix-xdp_rxq_info-leak.patch +drm-exynos-fimd-guard-display-clock-control-with-run.patch +spi-spi-qpic-snand-reallocate-bam-transactions.patch +spi-spi-fsl-dspi-clear-completion-counter-before-ini.patch +drm-i915-selftests-change-mock_request-to-return-err.patch +nvme-fix-incorrect-cdw15-value-in-passthru-error-log.patch +nvmet-fix-memory-leak-of-bio-integrity.patch +nvme-pci-refresh-visible-attrs-after-being-checked.patch +platform-x86-dell-wmi-sysman-fix-wmi-data-block-retr.patch +platform-x86-hp-bioscfg-fix-class-device-unregistrat.patch +platform-x86-think-lmi-fix-class-device-unregistrati.patch +platform-x86-dell-wmi-sysman-fix-class-device-unregi.patch +platform-mellanox-mlxreg-lc-fix-logic-error-in-power.patch +drm-bridge-panel-move-prepare_prev_first-handling-to.patch +drm-bridge-aux-hpd-bridge-fix-assignment-of-the-of_n.patch +smb-client-fix-warning-when-reconnecting-channel.patch +net-usb-lan78xx-fix-warn-in-__netif_napi_del_locked-.patch +drm-vmwgfx-fix-guests-running-with-tdx-sev.patch +drm-i915-gt-fix-timeline-left-held-on-vma-alloc-erro.patch +drm-i915-gsc-mei-interrupt-top-half-should-be-in-irq.patch +nvme-multipath-fix-suspicious-rcu-usage-warning.patch +idpf-return-0-size-for-rss-key-if-not-supported.patch +idpf-convert-control-queue-mutex-to-a-spinlock.patch +igc-disable-l1.2-pci-e-link-substate-to-avoid-perfor.patch +drm-xe-fix-out-of-bounds-field-write-in-mi_store_dat.patch +netfs-fix-hang-due-to-missing-case-in-final-dio-read.patch +netfs-fix-looping-in-wait-functions.patch +netfs-fix-ref-leak-on-inserted-extra-subreq-in-write.patch +smb-client-set-missing-retry-flag-in-smb2_writev_cal.patch +smb-client-set-missing-retry-flag-in-cifs_readv_call.patch +smb-client-set-missing-retry-flag-in-cifs_writev_cal.patch +netfs-fix-i_size-updating.patch +drm-xe-guc-enable-w-a-16026508708.patch +drm-xe-guc_pc-add-_locked-variant-for-min-max-freq.patch +drm-xe-split-xe_device_td_flush.patch +drm-xe-bmg-update-wa_14022085890.patch +drm-xe-bmg-update-wa_22019338487.patch +lib-test_objagg-set-error-message-in-check_expect_hi.patch +amd-xgbe-align-cl37-an-sequence-as-per-databook.patch +enic-fix-incorrect-mtu-comparison-in-enic_change_mtu.patch +rose-fix-dangling-neighbour-pointers-in-rose_rt_devi.patch +nui-fix-dma_mapping_error-check.patch +net-sched-always-pass-notifications-when-child-class.patch +amd-xgbe-do-not-double-read-link-status.patch +net-ipv4-fix-stat-increase-when-udp-early-demux-drop.patch +smb-client-fix-race-condition-in-negotiate-timeout-b.patch +smb-client-fix-native-smb-symlink-traversal.patch +netfs-fix-double-put-of-request.patch +drm-xe-allow-dropping-kunit-dependency-as-built-in.patch +x86-platform-amd-move-final-timeout-check-to-after-f.patch +drm-msm-fix-a-fence-leak-in-submit-error-path.patch +drm-msm-fix-another-leak-in-the-submit-error-path.patch +alsa-sb-don-t-allow-changing-the-dma-mode-during-ope.patch +alsa-sb-force-to-disable-dmas-once-when-dma-mode-is-.patch +ata-libata-acpi-do-not-assume-40-wire-cable-if-no-de.patch +ata-pata_cs5536-fix-build-on-32-bit-uml.patch +asoc-amd-yc-add-quirk-for-msi-bravo-17-d7vf-internal.patch +platform-x86-amd-pmc-add-pcspecialist-lafite-pro-v-1.patch +genirq-irq_sim-initialize-work-context-pointers-prop.patch +powerpc-fix-struct-termio-related-ioctl-macros.patch +asoc-amd-yc-update-quirk-data-for-hp-victus.patch +regulator-fan53555-add-enable_time-support-and-soft-.patch +scsi-target-fix-null-pointer-dereference-in-core_scs.patch +aoe-defer-rexmit-timer-downdev-work-to-workqueue.patch +wifi-mac80211-drop-invalid-source-address-ocb-frames.patch +wifi-ath6kl-remove-warn-on-bad-firmware-input.patch +acpica-refuse-to-evaluate-a-method-if-arguments-are-.patch +mtd-spinand-fix-memory-leak-of-ecc-engine-conf.patch +rcu-return-early-if-callback-is-not-specified.patch diff --git a/queue-6.15/smb-client-fix-native-smb-symlink-traversal.patch b/queue-6.15/smb-client-fix-native-smb-symlink-traversal.patch new file mode 100644 index 0000000000..69b8a7d144 --- /dev/null +++ b/queue-6.15/smb-client-fix-native-smb-symlink-traversal.patch @@ -0,0 +1,175 @@ +From c6ffb5b8f6c6c12f71c77531a4d86e3ffe99dfd9 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Thu, 3 Jul 2025 17:57:19 -0300 +Subject: smb: client: fix native SMB symlink traversal + +From: Paulo Alcantara + +[ Upstream commit 3363da82e02f1bddc54faa92ea430c6532e2cd2e ] + +We've seen customers having shares mounted in paths like /??/C:/ or +/??/UNC/foo.example.com/share in order to get their native SMB +symlinks successfully followed from different mounts. + +After commit 12b466eb52d9 ("cifs: Fix creating and resolving absolute NT-style symlinks"), +the client would then convert absolute paths from "/??/C:/" to "/mnt/c/" +by default. The absolute paths would vary depending on the value of +symlinkroot= mount option. + +Fix this by restoring old behavior of not trying to convert absolute +paths by default. Only do this if symlinkroot= was _explicitly_ set. + +Before patch: + + $ mount.cifs //w22-fs0/test2 /mnt/1 -o vers=3.1.1,username=xxx,password=yyy + $ ls -l /mnt/1/symlink2 + lrwxr-xr-x 1 root root 15 Jun 20 14:22 /mnt/1/symlink2 -> /mnt/c/testfile + $ mkdir -p /??/C:; echo foo > //??/C:/testfile + $ cat /mnt/1/symlink2 + cat: /mnt/1/symlink2: No such file or directory + +After patch: + + $ mount.cifs //w22-fs0/test2 /mnt/1 -o vers=3.1.1,username=xxx,password=yyy + $ ls -l /mnt/1/symlink2 + lrwxr-xr-x 1 root root 15 Jun 20 14:22 /mnt/1/symlink2 -> '/??/C:/testfile' + $ mkdir -p /??/C:; echo foo > //??/C:/testfile + $ cat /mnt/1/symlink2 + foo + +Cc: linux-cifs@vger.kernel.org +Reported-by: Pierguido Lambri +Cc: David Howells +Cc: Stefan Metzmacher +Fixes: 12b466eb52d9 ("cifs: Fix creating and resolving absolute NT-style symlinks") +Signed-off-by: Paulo Alcantara (Red Hat) +Signed-off-by: Steve French +Signed-off-by: Sasha Levin +--- + fs/smb/client/fs_context.c | 17 +++++++---------- + fs/smb/client/reparse.c | 22 +++++++++++++--------- + 2 files changed, 20 insertions(+), 19 deletions(-) + +diff --git a/fs/smb/client/fs_context.c b/fs/smb/client/fs_context.c +index a634a34d4086a..59ccc2229ab30 100644 +--- a/fs/smb/client/fs_context.c ++++ b/fs/smb/client/fs_context.c +@@ -1824,10 +1824,14 @@ static int smb3_fs_context_parse_param(struct fs_context *fc, + cifs_errorf(fc, "symlinkroot mount options must be absolute path\n"); + goto cifs_parse_mount_err; + } +- kfree(ctx->symlinkroot); +- ctx->symlinkroot = kstrdup(param->string, GFP_KERNEL); +- if (!ctx->symlinkroot) ++ if (strnlen(param->string, PATH_MAX) == PATH_MAX) { ++ cifs_errorf(fc, "symlinkroot path too long (max path length: %u)\n", ++ PATH_MAX - 1); + goto cifs_parse_mount_err; ++ } ++ kfree(ctx->symlinkroot); ++ ctx->symlinkroot = param->string; ++ param->string = NULL; + break; + } + /* case Opt_ignore: - is ignored as expected ... */ +@@ -1837,13 +1841,6 @@ static int smb3_fs_context_parse_param(struct fs_context *fc, + goto cifs_parse_mount_err; + } + +- /* +- * By default resolve all native absolute symlinks relative to "/mnt/". +- * Same default has drvfs driver running in WSL for resolving SMB shares. +- */ +- if (!ctx->symlinkroot) +- ctx->symlinkroot = kstrdup("/mnt/", GFP_KERNEL); +- + return 0; + + cifs_parse_mount_err: +diff --git a/fs/smb/client/reparse.c b/fs/smb/client/reparse.c +index 1c40e42e4d897..5fa29a97ac154 100644 +--- a/fs/smb/client/reparse.c ++++ b/fs/smb/client/reparse.c +@@ -57,6 +57,7 @@ static int create_native_symlink(const unsigned int xid, struct inode *inode, + struct reparse_symlink_data_buffer *buf = NULL; + struct cifs_open_info_data data = {}; + struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); ++ const char *symroot = cifs_sb->ctx->symlinkroot; + struct inode *new; + struct kvec iov; + __le16 *path = NULL; +@@ -82,7 +83,8 @@ static int create_native_symlink(const unsigned int xid, struct inode *inode, + .symlink_target = symlink_target, + }; + +- if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_POSIX_PATHS) && symname[0] == '/') { ++ if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_POSIX_PATHS) && ++ symroot && symname[0] == '/') { + /* + * This is a request to create an absolute symlink on the server + * which does not support POSIX paths, and expects symlink in +@@ -92,7 +94,7 @@ static int create_native_symlink(const unsigned int xid, struct inode *inode, + * ensure compatibility of this symlink stored in absolute form + * on the SMB server. + */ +- if (!strstarts(symname, cifs_sb->ctx->symlinkroot)) { ++ if (!strstarts(symname, symroot)) { + /* + * If the absolute Linux symlink target path is not + * inside "symlinkroot" location then there is no way +@@ -101,12 +103,12 @@ static int create_native_symlink(const unsigned int xid, struct inode *inode, + cifs_dbg(VFS, + "absolute symlink '%s' cannot be converted to NT format " + "because it is outside of symlinkroot='%s'\n", +- symname, cifs_sb->ctx->symlinkroot); ++ symname, symroot); + rc = -EINVAL; + goto out; + } +- len = strlen(cifs_sb->ctx->symlinkroot); +- if (cifs_sb->ctx->symlinkroot[len-1] != '/') ++ len = strlen(symroot); ++ if (symroot[len - 1] != '/') + len++; + if (symname[len] >= 'a' && symname[len] <= 'z' && + (symname[len+1] == '/' || symname[len+1] == '\0')) { +@@ -782,6 +784,7 @@ int smb2_parse_native_symlink(char **target, const char *buf, unsigned int len, + const char *full_path, + struct cifs_sb_info *cifs_sb) + { ++ const char *symroot = cifs_sb->ctx->symlinkroot; + char sep = CIFS_DIR_SEP(cifs_sb); + char *linux_target = NULL; + char *smb_target = NULL; +@@ -815,7 +818,8 @@ int smb2_parse_native_symlink(char **target, const char *buf, unsigned int len, + goto out; + } + +- if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_POSIX_PATHS) && !relative) { ++ if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_POSIX_PATHS) && ++ symroot && !relative) { + /* + * This is an absolute symlink from the server which does not + * support POSIX paths, so the symlink is in NT-style path. +@@ -907,15 +911,15 @@ int smb2_parse_native_symlink(char **target, const char *buf, unsigned int len, + } + + abs_path_len = strlen(abs_path)+1; +- symlinkroot_len = strlen(cifs_sb->ctx->symlinkroot); +- if (cifs_sb->ctx->symlinkroot[symlinkroot_len-1] == '/') ++ symlinkroot_len = strlen(symroot); ++ if (symroot[symlinkroot_len - 1] == '/') + symlinkroot_len--; + linux_target = kmalloc(symlinkroot_len + 1 + abs_path_len, GFP_KERNEL); + if (!linux_target) { + rc = -ENOMEM; + goto out; + } +- memcpy(linux_target, cifs_sb->ctx->symlinkroot, symlinkroot_len); ++ memcpy(linux_target, symroot, symlinkroot_len); + linux_target[symlinkroot_len] = '/'; + memcpy(linux_target + symlinkroot_len + 1, abs_path, abs_path_len); + } else if (smb_target[0] == sep && relative) { +-- +2.39.5 + diff --git a/queue-6.15/smb-client-fix-race-condition-in-negotiate-timeout-b.patch b/queue-6.15/smb-client-fix-race-condition-in-negotiate-timeout-b.patch new file mode 100644 index 0000000000..f831f55998 --- /dev/null +++ b/queue-6.15/smb-client-fix-race-condition-in-negotiate-timeout-b.patch @@ -0,0 +1,128 @@ +From e69b5754a5e3dbe6959fd67f76d9a77e27202064 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Thu, 3 Jul 2025 21:29:52 +0800 +Subject: smb: client: fix race condition in negotiate timeout by using more + precise timing + +From: Wang Zhaolong + +[ Upstream commit 266b5d02e14f3a0e07414e11f239397de0577a1d ] + +When the SMB server reboots and the client immediately accesses the mount +point, a race condition can occur that causes operations to fail with +"Host is down" error. + +Reproduction steps: + # Mount SMB share + mount -t cifs //192.168.245.109/TEST /mnt/ -o xxxx + ls /mnt + + # Reboot server + ssh root@192.168.245.109 reboot + ssh root@192.168.245.109 /path/to/cifs_server_setup.sh + ssh root@192.168.245.109 systemctl stop firewalld + + # Immediate access fails + ls /mnt + ls: cannot access '/mnt': Host is down + + # But works if there is a delay + +The issue is caused by a race condition between negotiate and reconnect. +The 20-second negotiate timeout mechanism can interfere with the normal +recovery process when both are triggered simultaneously. + + ls cifsd +--------------------------------------------------- + cifs_getattr + cifs_revalidate_dentry + cifs_get_inode_info + cifs_get_fattr + smb2_query_path_info + smb2_compound_op + SMB2_open_init + smb2_reconnect + cifs_negotiate_protocol + smb2_negotiate + cifs_send_recv + smb_send_rqst + wait_for_response + cifs_demultiplex_thread + cifs_read_from_socket + cifs_readv_from_socket + server_unresponsive + cifs_reconnect + __cifs_reconnect + cifs_abort_connection + mid->mid_state = MID_RETRY_NEEDED + cifs_wake_up_task + cifs_sync_mid_result + // case MID_RETRY_NEEDED + rc = -EAGAIN; + // In smb2_negotiate() + rc = -EHOSTDOWN; + +The server_unresponsive() timeout triggers cifs_reconnect(), which aborts +ongoing mid requests and causes the ls command to receive -EAGAIN, leading +to -EHOSTDOWN. + +Fix this by introducing a dedicated `neg_start` field to +precisely tracks when the negotiate process begins. The timeout check +now uses this accurate timestamp instead of `lstrp`, ensuring that: + +1. Timeout is only triggered after negotiate has actually run for 20s +2. The mechanism doesn't interfere with concurrent recovery processes +3. Uninitialized timestamps (value 0) don't trigger false timeouts + +Fixes: 7ccc1465465d ("smb: client: fix hang in wait_for_response() for negproto") +Signed-off-by: Wang Zhaolong +Signed-off-by: Steve French +Signed-off-by: Sasha Levin +--- + fs/smb/client/cifsglob.h | 1 + + fs/smb/client/connect.c | 7 ++++--- + 2 files changed, 5 insertions(+), 3 deletions(-) + +diff --git a/fs/smb/client/cifsglob.h b/fs/smb/client/cifsglob.h +index b89db44e0a8e3..eed29f043114a 100644 +--- a/fs/smb/client/cifsglob.h ++++ b/fs/smb/client/cifsglob.h +@@ -777,6 +777,7 @@ struct TCP_Server_Info { + __le32 session_key_id; /* retrieved from negotiate response and send in session setup request */ + struct session_key session_key; + unsigned long lstrp; /* when we got last response from this server */ ++ unsigned long neg_start; /* when negotiate started (jiffies) */ + struct cifs_secmech secmech; /* crypto sec mech functs, descriptors */ + #define CIFS_NEGFLAVOR_UNENCAP 1 /* wct == 17, but no ext_sec */ + #define CIFS_NEGFLAVOR_EXTENDED 2 /* wct == 17, ext_sec bit set */ +diff --git a/fs/smb/client/connect.c b/fs/smb/client/connect.c +index e92c7b71626fd..b4b162c93106b 100644 +--- a/fs/smb/client/connect.c ++++ b/fs/smb/client/connect.c +@@ -679,12 +679,12 @@ server_unresponsive(struct TCP_Server_Info *server) + /* + * If we're in the process of mounting a share or reconnecting a session + * and the server abruptly shut down (e.g. socket wasn't closed, packet +- * had been ACK'ed but no SMB response), don't wait longer than 20s to +- * negotiate protocol. ++ * had been ACK'ed but no SMB response), don't wait longer than 20s from ++ * when negotiate actually started. + */ + spin_lock(&server->srv_lock); + if (server->tcpStatus == CifsInNegotiate && +- time_after(jiffies, server->lstrp + 20 * HZ)) { ++ time_after(jiffies, server->neg_start + 20 * HZ)) { + spin_unlock(&server->srv_lock); + cifs_reconnect(server, false); + return true; +@@ -4209,6 +4209,7 @@ cifs_negotiate_protocol(const unsigned int xid, struct cifs_ses *ses, + + server->lstrp = jiffies; + server->tcpStatus = CifsInNegotiate; ++ server->neg_start = jiffies; + spin_unlock(&server->srv_lock); + + rc = server->ops->negotiate(xid, ses, server); +-- +2.39.5 + diff --git a/queue-6.15/smb-client-fix-warning-when-reconnecting-channel.patch b/queue-6.15/smb-client-fix-warning-when-reconnecting-channel.patch new file mode 100644 index 0000000000..de73f2b4e2 --- /dev/null +++ b/queue-6.15/smb-client-fix-warning-when-reconnecting-channel.patch @@ -0,0 +1,129 @@ +From a7e0338cabed78ba10808f13b1f2d0c45e153769 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Wed, 25 Jun 2025 00:00:17 -0300 +Subject: smb: client: fix warning when reconnecting channel + +From: Paulo Alcantara + +[ Upstream commit 3bbe46716092d8ef6b0df4b956f585c5cd0fc78e ] + +When reconnecting a channel in smb2_reconnect_server(), a dummy tcon +is passed down to smb2_reconnect() with ->query_interface +uninitialized, so we can't call queue_delayed_work() on it. + +Fix the following warning by ensuring that we're queueing the delayed +worker from correct tcon. + +WARNING: CPU: 4 PID: 1126 at kernel/workqueue.c:2498 __queue_delayed_work+0x1d2/0x200 +Modules linked in: cifs cifs_arc4 nls_ucs2_utils cifs_md4 [last unloaded: cifs] +CPU: 4 UID: 0 PID: 1126 Comm: kworker/4:0 Not tainted 6.16.0-rc3 #5 PREEMPT(voluntary) +Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.3-4.fc42 04/01/2014 +Workqueue: cifsiod smb2_reconnect_server [cifs] +RIP: 0010:__queue_delayed_work+0x1d2/0x200 +Code: 41 5e 41 5f e9 7f ee ff ff 90 0f 0b 90 e9 5d ff ff ff bf 02 00 +00 00 e8 6c f3 07 00 89 c3 eb bd 90 0f 0b 90 e9 57 f> 0b 90 e9 65 fe +ff ff 90 0f 0b 90 e9 72 fe ff ff 90 0f 0b 90 e9 +RSP: 0018:ffffc900014afad8 EFLAGS: 00010003 +RAX: 0000000000000000 RBX: ffff888124d99988 RCX: ffffffff81399cc1 +RDX: dffffc0000000000 RSI: ffff888114326e00 RDI: ffff888124d999f0 +RBP: 000000000000ea60 R08: 0000000000000001 R09: ffffed10249b3331 +R10: ffff888124d9998f R11: 0000000000000004 R12: 0000000000000040 +R13: ffff888114326e00 R14: ffff888124d999d8 R15: ffff888114939020 +FS: 0000000000000000(0000) GS:ffff88829f7fe000(0000) knlGS:0000000000000000 +CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 +CR2: 00007ffe7a2b4038 CR3: 0000000120a6f000 CR4: 0000000000750ef0 +PKRU: 55555554 +Call Trace: + + queue_delayed_work_on+0xb4/0xc0 + smb2_reconnect+0xb22/0xf50 [cifs] + smb2_reconnect_server+0x413/0xd40 [cifs] + ? __pfx_smb2_reconnect_server+0x10/0x10 [cifs] + ? local_clock_noinstr+0xd/0xd0 + ? local_clock+0x15/0x30 + ? lock_release+0x29b/0x390 + process_one_work+0x4c5/0xa10 + ? __pfx_process_one_work+0x10/0x10 + ? __list_add_valid_or_report+0x37/0x120 + worker_thread+0x2f1/0x5a0 + ? __kthread_parkme+0xde/0x100 + ? __pfx_worker_thread+0x10/0x10 + kthread+0x1fe/0x380 + ? kthread+0x10f/0x380 + ? __pfx_kthread+0x10/0x10 + ? local_clock_noinstr+0xd/0xd0 + ? ret_from_fork+0x1b/0x1f0 + ? local_clock+0x15/0x30 + ? lock_release+0x29b/0x390 + ? rcu_is_watching+0x20/0x50 + ? __pfx_kthread+0x10/0x10 + ret_from_fork+0x15b/0x1f0 + ? __pfx_kthread+0x10/0x10 + ret_from_fork_asm+0x1a/0x30 + +irq event stamp: 1116206 +hardirqs last enabled at (1116205): [] __up_console_sem+0x52/0x60 +hardirqs last disabled at (1116206): [] queue_delayed_work_on+0x6e/0xc0 +softirqs last enabled at (1116138): [] __smb_send_rqst+0x42d/0x950 [cifs] +softirqs last disabled at (1116136): [] release_sock+0x21/0xf0 + +Cc: linux-cifs@vger.kernel.org +Reported-by: David Howells +Fixes: 42ca547b13a2 ("cifs: do not disable interface polling on failure") +Reviewed-by: David Howells +Tested-by: David Howells +Reviewed-by: Shyam Prasad N +Signed-off-by: Paulo Alcantara (Red Hat) +Signed-off-by: David Howells +Tested-by: Steve French +Signed-off-by: Steve French +Signed-off-by: Sasha Levin +--- + fs/smb/client/cifsglob.h | 1 + + fs/smb/client/smb2pdu.c | 10 ++++------ + 2 files changed, 5 insertions(+), 6 deletions(-) + +diff --git a/fs/smb/client/cifsglob.h b/fs/smb/client/cifsglob.h +index 56381cbb63990..b89db44e0a8e3 100644 +--- a/fs/smb/client/cifsglob.h ++++ b/fs/smb/client/cifsglob.h +@@ -1303,6 +1303,7 @@ struct cifs_tcon { + bool use_persistent:1; /* use persistent instead of durable handles */ + bool no_lease:1; /* Do not request leases on files or directories */ + bool use_witness:1; /* use witness protocol */ ++ bool dummy:1; /* dummy tcon used for reconnecting channels */ + __le32 capabilities; + __u32 share_flags; + __u32 maximal_access; +diff --git a/fs/smb/client/smb2pdu.c b/fs/smb/client/smb2pdu.c +index 72903265b1706..c3b212175b2bf 100644 +--- a/fs/smb/client/smb2pdu.c ++++ b/fs/smb/client/smb2pdu.c +@@ -423,9 +423,9 @@ smb2_reconnect(__le16 smb2_command, struct cifs_tcon *tcon, + free_xid(xid); + ses->flags &= ~CIFS_SES_FLAGS_PENDING_QUERY_INTERFACES; + +- /* regardless of rc value, setup polling */ +- queue_delayed_work(cifsiod_wq, &tcon->query_interfaces, +- (SMB_INTERFACE_POLL_INTERVAL * HZ)); ++ if (!tcon->ipc && !tcon->dummy) ++ queue_delayed_work(cifsiod_wq, &tcon->query_interfaces, ++ (SMB_INTERFACE_POLL_INTERVAL * HZ)); + + mutex_unlock(&ses->session_mutex); + +@@ -4221,10 +4221,8 @@ void smb2_reconnect_server(struct work_struct *work) + } + goto done; + } +- + tcon->status = TID_GOOD; +- tcon->retry = false; +- tcon->need_reconnect = false; ++ tcon->dummy = true; + + /* now reconnect sessions for necessary channels */ + list_for_each_entry_safe(ses, ses2, &tmp_ses_list, rlist) { +-- +2.39.5 + diff --git a/queue-6.15/smb-client-set-missing-retry-flag-in-cifs_readv_call.patch b/queue-6.15/smb-client-set-missing-retry-flag-in-cifs_readv_call.patch new file mode 100644 index 0000000000..4b4bf33f93 --- /dev/null +++ b/queue-6.15/smb-client-set-missing-retry-flag-in-cifs_readv_call.patch @@ -0,0 +1,40 @@ +From e7c29937335f8a22a6a0627fc23a1a8c3987a63f Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Tue, 1 Jul 2025 17:38:42 +0100 +Subject: smb: client: set missing retry flag in cifs_readv_callback() + +From: Paulo Alcantara + +[ Upstream commit 0e60bae24ad28ab06a485698077d3c626f1e54ab ] + +Set NETFS_SREQ_NEED_RETRY flag to tell netfslib that the subreq needs +to be retried. + +Fixes: ee4cdf7ba857 ("netfs: Speed up buffered reading") +Signed-off-by: Paulo Alcantara (Red Hat) +Signed-off-by: David Howells +Link: https://lore.kernel.org/20250701163852.2171681-8-dhowells@redhat.com +Tested-by: Steve French +Cc: linux-cifs@vger.kernel.org +Cc: netfs@lists.linux.dev +Signed-off-by: Christian Brauner +Signed-off-by: Sasha Levin +--- + fs/smb/client/cifssmb.c | 1 + + 1 file changed, 1 insertion(+) + +diff --git a/fs/smb/client/cifssmb.c b/fs/smb/client/cifssmb.c +index 7216fcec79e8b..f9ccae5de5b88 100644 +--- a/fs/smb/client/cifssmb.c ++++ b/fs/smb/client/cifssmb.c +@@ -1335,6 +1335,7 @@ cifs_readv_callback(struct mid_q_entry *mid) + break; + case MID_REQUEST_SUBMITTED: + case MID_RETRY_NEEDED: ++ __set_bit(NETFS_SREQ_NEED_RETRY, &rdata->subreq.flags); + rdata->result = -EAGAIN; + if (server->sign && rdata->got_bytes) + /* reset bytes number since we can not check a sign */ +-- +2.39.5 + diff --git a/queue-6.15/smb-client-set-missing-retry-flag-in-cifs_writev_cal.patch b/queue-6.15/smb-client-set-missing-retry-flag-in-cifs_writev_cal.patch new file mode 100644 index 0000000000..4878210803 --- /dev/null +++ b/queue-6.15/smb-client-set-missing-retry-flag-in-cifs_writev_cal.patch @@ -0,0 +1,40 @@ +From 19af7c76cda1f124106511b9989271fa979620e2 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Tue, 1 Jul 2025 17:38:43 +0100 +Subject: smb: client: set missing retry flag in cifs_writev_callback() + +From: Paulo Alcantara + +[ Upstream commit 74ee76bea4b445c023d04806e0bcd78a912fd30b ] + +Set NETFS_SREQ_NEED_RETRY flag to tell netfslib that the subreq needs +to be retried. + +Fixes: ee4cdf7ba857 ("netfs: Speed up buffered reading") +Signed-off-by: Paulo Alcantara (Red Hat) +Signed-off-by: David Howells +Link: https://lore.kernel.org/20250701163852.2171681-9-dhowells@redhat.com +Tested-by: Steve French +Cc: linux-cifs@vger.kernel.org +Cc: netfs@lists.linux.dev +Signed-off-by: Christian Brauner +Signed-off-by: Sasha Levin +--- + fs/smb/client/cifssmb.c | 1 + + 1 file changed, 1 insertion(+) + +diff --git a/fs/smb/client/cifssmb.c b/fs/smb/client/cifssmb.c +index f9ccae5de5b88..0e509a0433fb6 100644 +--- a/fs/smb/client/cifssmb.c ++++ b/fs/smb/client/cifssmb.c +@@ -1715,6 +1715,7 @@ cifs_writev_callback(struct mid_q_entry *mid) + break; + case MID_REQUEST_SUBMITTED: + case MID_RETRY_NEEDED: ++ __set_bit(NETFS_SREQ_NEED_RETRY, &wdata->subreq.flags); + result = -EAGAIN; + break; + default: +-- +2.39.5 + diff --git a/queue-6.15/smb-client-set-missing-retry-flag-in-smb2_writev_cal.patch b/queue-6.15/smb-client-set-missing-retry-flag-in-smb2_writev_cal.patch new file mode 100644 index 0000000000..386ed8aafe --- /dev/null +++ b/queue-6.15/smb-client-set-missing-retry-flag-in-smb2_writev_cal.patch @@ -0,0 +1,40 @@ +From 80bcde21959a8a7dd7e7312443f7e90b571ceb3f Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Tue, 1 Jul 2025 17:38:41 +0100 +Subject: smb: client: set missing retry flag in smb2_writev_callback() + +From: Paulo Alcantara + +[ Upstream commit e67e75edeb88022c04f8e0a173e1ff6dc688f155 ] + +Set NETFS_SREQ_NEED_RETRY flag to tell netfslib that the subreq needs +to be retried. + +Fixes: ee4cdf7ba857 ("netfs: Speed up buffered reading") +Signed-off-by: Paulo Alcantara (Red Hat) +Signed-off-by: David Howells +Link: https://lore.kernel.org/20250701163852.2171681-7-dhowells@redhat.com +Tested-by: Steve French +Cc: linux-cifs@vger.kernel.org +Cc: netfs@lists.linux.dev +Signed-off-by: Christian Brauner +Signed-off-by: Sasha Levin +--- + fs/smb/client/smb2pdu.c | 1 + + 1 file changed, 1 insertion(+) + +diff --git a/fs/smb/client/smb2pdu.c b/fs/smb/client/smb2pdu.c +index c3b212175b2bf..2c0cc544dfb31 100644 +--- a/fs/smb/client/smb2pdu.c ++++ b/fs/smb/client/smb2pdu.c +@@ -4852,6 +4852,7 @@ smb2_writev_callback(struct mid_q_entry *mid) + break; + case MID_REQUEST_SUBMITTED: + case MID_RETRY_NEEDED: ++ __set_bit(NETFS_SREQ_NEED_RETRY, &wdata->subreq.flags); + result = -EAGAIN; + break; + case MID_RESPONSE_MALFORMED: +-- +2.39.5 + diff --git a/queue-6.15/spi-spi-fsl-dspi-clear-completion-counter-before-ini.patch b/queue-6.15/spi-spi-fsl-dspi-clear-completion-counter-before-ini.patch new file mode 100644 index 0000000000..a812b9c37a --- /dev/null +++ b/queue-6.15/spi-spi-fsl-dspi-clear-completion-counter-before-ini.patch @@ -0,0 +1,61 @@ +From c87f2eeea1991258a66d2d924907494e9bf2d131 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Fri, 27 Jun 2025 11:21:37 +0100 +Subject: spi: spi-fsl-dspi: Clear completion counter before initiating + transfer + +From: James Clark + +[ Upstream commit fa60c094c19b97e103d653f528f8d9c178b6a5f5 ] + +In target mode, extra interrupts can be received between the end of a +transfer and halting the module if the host continues sending more data. +If the interrupt from this occurs after the reinit_completion() then the +completion counter is left at a non-zero value. The next unrelated +transfer initiated by userspace will then complete immediately without +waiting for the interrupt or writing to the RX buffer. + +Fix it by resetting the counter before the transfer so that lingering +values are cleared. This is done after clearing the FIFOs and the +status register but before the transfer is initiated, so no interrupts +should be received at this point resulting in other race conditions. + +Fixes: 4f5ee75ea171 ("spi: spi-fsl-dspi: Replace interruptible wait queue with a simple completion") +Signed-off-by: James Clark +Reviewed-by: Frank Li +Link: https://patch.msgid.link/20250627-james-nxp-spi-dma-v4-1-178dba20c120@linaro.org +Signed-off-by: Mark Brown +Signed-off-by: Sasha Levin +--- + drivers/spi/spi-fsl-dspi.c | 11 ++++++++++- + 1 file changed, 10 insertions(+), 1 deletion(-) + +diff --git a/drivers/spi/spi-fsl-dspi.c b/drivers/spi/spi-fsl-dspi.c +index 863781ba6c160..0dcd491140950 100644 +--- a/drivers/spi/spi-fsl-dspi.c ++++ b/drivers/spi/spi-fsl-dspi.c +@@ -983,11 +983,20 @@ static int dspi_transfer_one_message(struct spi_controller *ctlr, + if (dspi->devtype_data->trans_mode == DSPI_DMA_MODE) { + status = dspi_dma_xfer(dspi); + } else { ++ /* ++ * Reinitialize the completion before transferring data ++ * to avoid the case where it might remain in the done ++ * state due to a spurious interrupt from a previous ++ * transfer. This could falsely signal that the current ++ * transfer has completed. ++ */ ++ if (dspi->irq) ++ reinit_completion(&dspi->xfer_done); ++ + dspi_fifo_write(dspi); + + if (dspi->irq) { + wait_for_completion(&dspi->xfer_done); +- reinit_completion(&dspi->xfer_done); + } else { + do { + status = dspi_poll(dspi); +-- +2.39.5 + diff --git a/queue-6.15/spi-spi-qpic-snand-reallocate-bam-transactions.patch b/queue-6.15/spi-spi-qpic-snand-reallocate-bam-transactions.patch new file mode 100644 index 0000000000..3850586bec --- /dev/null +++ b/queue-6.15/spi-spi-qpic-snand-reallocate-bam-transactions.patch @@ -0,0 +1,91 @@ +From 568dfe17798b5b1d3677028f2839d3270548d387 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Wed, 18 Jun 2025 22:22:49 +0200 +Subject: spi: spi-qpic-snand: reallocate BAM transactions + +From: Gabor Juhos + +[ Upstream commit d85d0380292a7e618915069c3579ae23c7c80339 ] + +Using the mtd_nandbiterrs module for testing the driver occasionally +results in weird things like below. + +1. swiotlb mapping fails with the following message: + + [ 85.926216] qcom_snand 79b0000.spi: swiotlb buffer is full (sz: 4294967294 bytes), total 512 (slots), used 0 (slots) + [ 85.932937] qcom_snand 79b0000.spi: failure in mapping desc + [ 87.999314] qcom_snand 79b0000.spi: failure to write raw page + [ 87.999352] mtd_nandbiterrs: error: write_oob failed (-110) + + Rebooting the board after this causes a panic due to a NULL pointer + dereference. + +2. If the swiotlb mapping does not fail, rebooting the board may result + in a different panic due to a bad spinlock magic: + + [ 256.104459] BUG: spinlock bad magic on CPU#3, procd/2241 + [ 256.104488] Unable to handle kernel paging request at virtual address ffffffff0000049b + ... + +Investigating the issue revealed that these symptoms are results of +memory corruption which is caused by out of bounds access within the +driver. + +The driver uses a dynamically allocated structure for BAM transactions, +which structure must have enough space for all possible variations of +different flash operations initiated by the driver. The required space +heavily depends on the actual number of 'codewords' which is calculated +from the pagesize of the actual NAND chip. + +Although the qcom_nandc_alloc() function allocates memory for the BAM +transactions during probe, but since the actual number of 'codewords' +is not yet know the allocation is done for one 'codeword' only. + +Because of this, whenever the driver does a flash operation, and the +number of the required transactions exceeds the size of the allocated +arrays the driver accesses memory out of the allocated range. + +To avoid this, change the code to free the initially allocated BAM +transactions memory, and allocate a new one once the actual number of +'codewords' required for a given NAND chip is known. + +Fixes: 7304d1909080 ("spi: spi-qpic: add driver for QCOM SPI NAND flash Interface") +Reviewed-by: Md Sadre Alam +Signed-off-by: Gabor Juhos +Link: https://patch.msgid.link/20250618-qpic-snand-avoid-mem-corruption-v3-1-319c71296cda@gmail.com +Signed-off-by: Mark Brown +Signed-off-by: Sasha Levin +--- + drivers/spi/spi-qpic-snand.c | 16 ++++++++++++++++ + 1 file changed, 16 insertions(+) + +diff --git a/drivers/spi/spi-qpic-snand.c b/drivers/spi/spi-qpic-snand.c +index d3a4e091dca4e..80856d2fa35c6 100644 +--- a/drivers/spi/spi-qpic-snand.c ++++ b/drivers/spi/spi-qpic-snand.c +@@ -316,6 +316,22 @@ static int qcom_spi_ecc_init_ctx_pipelined(struct nand_device *nand) + + mtd_set_ooblayout(mtd, &qcom_spi_ooblayout); + ++ /* ++ * Free the temporary BAM transaction allocated initially by ++ * qcom_nandc_alloc(), and allocate a new one based on the ++ * updated max_cwperpage value. ++ */ ++ qcom_free_bam_transaction(snandc); ++ ++ snandc->max_cwperpage = cwperpage; ++ ++ snandc->bam_txn = qcom_alloc_bam_transaction(snandc); ++ if (!snandc->bam_txn) { ++ dev_err(snandc->dev, "failed to allocate BAM transaction\n"); ++ ret = -ENOMEM; ++ goto err_free_ecc_cfg; ++ } ++ + ecc_cfg->cfg0 = FIELD_PREP(CW_PER_PAGE_MASK, (cwperpage - 1)) | + FIELD_PREP(UD_SIZE_BYTES_MASK, ecc_cfg->cw_data) | + FIELD_PREP(DISABLE_STATUS_AFTER_WRITE, 1) | +-- +2.39.5 + diff --git a/queue-6.15/wifi-ath6kl-remove-warn-on-bad-firmware-input.patch b/queue-6.15/wifi-ath6kl-remove-warn-on-bad-firmware-input.patch new file mode 100644 index 0000000000..f7b60e15bc --- /dev/null +++ b/queue-6.15/wifi-ath6kl-remove-warn-on-bad-firmware-input.patch @@ -0,0 +1,43 @@ +From 0efb1ddf2bed00c7158c59778d15a415341bb441 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Tue, 17 Jun 2025 11:45:29 +0200 +Subject: wifi: ath6kl: remove WARN on bad firmware input + +From: Johannes Berg + +[ Upstream commit e7417421d89358da071fd2930f91e67c7128fbff ] + +If the firmware gives bad input, that's nothing to do with +the driver's stack at this point etc., so the WARN_ON() +doesn't add any value. Additionally, this is one of the +top syzbot reports now. Just print a message, and as an +added bonus, print the sizes too. + +Reported-by: syzbot+92c6dd14aaa230be6855@syzkaller.appspotmail.com +Tested-by: syzbot+92c6dd14aaa230be6855@syzkaller.appspotmail.com +Acked-by: Jeff Johnson +Link: https://patch.msgid.link/20250617114529.031a677a348e.I58bf1eb4ac16a82c546725ff010f3f0d2b0cca49@changeid +Signed-off-by: Johannes Berg +Signed-off-by: Sasha Levin +--- + drivers/net/wireless/ath/ath6kl/bmi.c | 4 +++- + 1 file changed, 3 insertions(+), 1 deletion(-) + +diff --git a/drivers/net/wireless/ath/ath6kl/bmi.c b/drivers/net/wireless/ath/ath6kl/bmi.c +index af98e871199d3..5a9e93fd1ef42 100644 +--- a/drivers/net/wireless/ath/ath6kl/bmi.c ++++ b/drivers/net/wireless/ath/ath6kl/bmi.c +@@ -87,7 +87,9 @@ int ath6kl_bmi_get_target_info(struct ath6kl *ar, + * We need to do some backwards compatibility to make this work. + */ + if (le32_to_cpu(targ_info->byte_count) != sizeof(*targ_info)) { +- WARN_ON(1); ++ ath6kl_err("mismatched byte count %d vs. expected %zd\n", ++ le32_to_cpu(targ_info->byte_count), ++ sizeof(*targ_info)); + return -EINVAL; + } + +-- +2.39.5 + diff --git a/queue-6.15/wifi-mac80211-drop-invalid-source-address-ocb-frames.patch b/queue-6.15/wifi-mac80211-drop-invalid-source-address-ocb-frames.patch new file mode 100644 index 0000000000..d2d872cad2 --- /dev/null +++ b/queue-6.15/wifi-mac80211-drop-invalid-source-address-ocb-frames.patch @@ -0,0 +1,42 @@ +From f58f0827bdd1c40b959ec2f8296c88b8eb317cb8 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Mon, 16 Jun 2025 17:18:38 +0200 +Subject: wifi: mac80211: drop invalid source address OCB frames + +From: Johannes Berg + +[ Upstream commit d1b1a5eb27c4948e8811cf4dbb05aaf3eb10700c ] + +In OCB, don't accept frames from invalid source addresses +(and in particular don't try to create stations for them), +drop the frames instead. + +Reported-by: syzbot+8b512026a7ec10dcbdd9@syzkaller.appspotmail.com +Closes: https://lore.kernel.org/r/6788d2d9.050a0220.20d369.0028.GAE@google.com/ +Signed-off-by: Johannes Berg +Tested-by: syzbot+8b512026a7ec10dcbdd9@syzkaller.appspotmail.com +Link: https://patch.msgid.link/20250616171838.7433379cab5d.I47444d63c72a0bd58d2e2b67bb99e1fea37eec6f@changeid +Signed-off-by: Johannes Berg +Signed-off-by: Sasha Levin +--- + net/mac80211/rx.c | 4 ++++ + 1 file changed, 4 insertions(+) + +diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c +index 09beb65d6108b..e73431549ce77 100644 +--- a/net/mac80211/rx.c ++++ b/net/mac80211/rx.c +@@ -4432,6 +4432,10 @@ static bool ieee80211_accept_frame(struct ieee80211_rx_data *rx) + if (!multicast && + !ether_addr_equal(sdata->dev->dev_addr, hdr->addr1)) + return false; ++ /* reject invalid/our STA address */ ++ if (!is_valid_ether_addr(hdr->addr2) || ++ ether_addr_equal(sdata->dev->dev_addr, hdr->addr2)) ++ return false; + if (!rx->sta) { + int rate_idx; + if (status->encoding != RX_ENC_LEGACY) +-- +2.39.5 + diff --git a/queue-6.15/x86-platform-amd-move-final-timeout-check-to-after-f.patch b/queue-6.15/x86-platform-amd-move-final-timeout-check-to-after-f.patch new file mode 100644 index 0000000000..a6c368ebd8 --- /dev/null +++ b/queue-6.15/x86-platform-amd-move-final-timeout-check-to-after-f.patch @@ -0,0 +1,61 @@ +From 2568d51844897d6b30aaa74c2824f4109413490e Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Thu, 5 Jun 2025 19:09:26 +0100 +Subject: x86/platform/amd: move final timeout check to after final sleep +MIME-Version: 1.0 +Content-Type: text/plain; charset=UTF-8 +Content-Transfer-Encoding: 8bit + +From: Jake Hillion + +[ Upstream commit f8afb12a2d7503de6558c23cacd7acbf6e9fe678 ] + +__hsmp_send_message sleeps between result read attempts and has a +timeout of 100ms. Under extreme load it's possible for these sleeps to +take a long time, exceeding the 100ms. In this case the current code +does not check the register and fails with ETIMEDOUT. + +Refactor the loop to ensure there is at least one read of the register +after a sleep of any duration. This removes instances of ETIMEDOUT with +a single caller, even with a misbehaving scheduler. Tested on AMD +Bergamo machines. + +Suggested-by: Blaise Sanouillet +Reviewed-by: Suma Hegde +Tested-by: Suma Hegde +Signed-off-by: Jake Hillion +Link: https://lore.kernel.org/r/20250605-amd-hsmp-v2-1-a811bc3dd74a@hillion.co.uk +Reviewed-by: Ilpo Järvinen +Signed-off-by: Ilpo Järvinen +Signed-off-by: Sasha Levin +--- + drivers/platform/x86/amd/hsmp/hsmp.c | 6 +++++- + 1 file changed, 5 insertions(+), 1 deletion(-) + +diff --git a/drivers/platform/x86/amd/hsmp/hsmp.c b/drivers/platform/x86/amd/hsmp/hsmp.c +index a3ac09a90de45..ab877112f4c80 100644 +--- a/drivers/platform/x86/amd/hsmp/hsmp.c ++++ b/drivers/platform/x86/amd/hsmp/hsmp.c +@@ -99,7 +99,7 @@ static int __hsmp_send_message(struct hsmp_socket *sock, struct hsmp_message *ms + short_sleep = jiffies + msecs_to_jiffies(HSMP_SHORT_SLEEP); + timeout = jiffies + msecs_to_jiffies(HSMP_MSG_TIMEOUT); + +- while (time_before(jiffies, timeout)) { ++ while (true) { + ret = sock->amd_hsmp_rdwr(sock, mbinfo->msg_resp_off, &mbox_status, HSMP_RD); + if (ret) { + dev_err(sock->dev, "Error %d reading mailbox status\n", ret); +@@ -108,6 +108,10 @@ static int __hsmp_send_message(struct hsmp_socket *sock, struct hsmp_message *ms + + if (mbox_status != HSMP_STATUS_NOT_READY) + break; ++ ++ if (!time_before(jiffies, timeout)) ++ break; ++ + if (time_before(jiffies, short_sleep)) + usleep_range(50, 100); + else +-- +2.39.5 +